In this paper, we introduce an end-to-end machine learning-based system for classifying autism spectrum disorder (ASD) using facial attributes such as expressions, action units, arousal, and valence. Our system classifies ASD using representations of different facial attributes from convolutional neural networks, which are trained on images in the wild. Our experimental results show that different facial attributes used in our system are statistically significant and improve sensitivity, specificity, and F1 score of ASD classification by a large margin. In particular, the addition of different facial attributes improves the performance of ASD classification by about 7% which achieves a F1 score of 76%.
http://arxiv.org/abs/1904.03616
Recovering a photorealistic face from an artistic portrait is a challenging task since crucial facial details are often distorted or completely lost in artistic compositions. To handle this loss, we propose an Attribute-guided Face Recovery from Portraits (AFRP) that utilizes a Face Recovery Network (FRN) and a Discriminative Network (DN). FRN consists of an autoencoder with residual block-embedded skip-connections and incorporates facial attribute vectors into the feature maps of input portraits at the bottleneck of the autoencoder. DN has multiple convolutional and fully-connected layers, and its role is to enforce FRN to generate authentic face images with corresponding facial attributes dictated by the input attribute vectors. %Leveraging on the spatial transformer networks, FRN automatically compensates for misalignments of portraits. % and generates aligned face images. For the preservation of identities, we impose the recovered and ground-truth faces to share similar visual features. Specifically, DN determines whether the recovered image looks like a real face and checks if the facial attributes extracted from the recovered image are consistent with given attributes. %Our method can recover high-quality photorealistic faces from unaligned portraits while preserving the identity of the face images as well as it can reconstruct a photorealistic face image with a desired set of attributes. Our method can recover photorealistic identity-preserving faces with desired attributes from unseen stylized portraits, artistic paintings, and hand-drawn sketches. On large-scale synthesized and sketch datasets, we demonstrate that our face recovery method achieves state-of-the-art results.
http://arxiv.org/abs/1904.03612
Given an artistic portrait, recovering the latent photorealistic face that preserves the subject’s identity is challenging because the facial details are often distorted or fully lost in artistic portraits. We develop an Identity-preserving Face Recovery from Portraits (IFRP) method that utilizes a Style Removal network (SRN) and a Discriminative Network (DN). Our SRN, composed of an autoencoder with residual block-embedded skip connections, is designed to transfer feature maps of stylized images to the feature maps of the corresponding photorealistic faces. Owing to the Spatial Transformer Network (STN), SRN automatically compensates for misalignments of stylized portraits to output aligned realistic face images. To ensure the identity preservation, we promote the recovered and ground truth faces to share similar visual features via a distance measure which compares features of recovered and ground truth faces extracted from a pre-trained FaceNet network. DN has multiple convolutional and fully-connected layers, and its role is to enforce recovered faces to be similar to authentic faces. Thus, we can recover high-quality photorealistic faces from unaligned portraits while preserving the identity of the face in an image. By conducting extensive evaluations on a large-scale synthesized dataset and a hand-drawn sketch dataset, we demonstrate that our method achieves superior face recovery and attains state-of-the-art results. In addition, our method can recover photorealistic faces from unseen stylized portraits, artistic paintings, and hand-drawn sketches.
http://arxiv.org/abs/1904.04241
Abnormal event detection in video is a challenging vision problem. Most existing approaches formulate abnormal event detection as an outlier detection task, due to the scarcity of anomalous data during training. Because of the lack of prior information regarding abnormal events, these methods are not fully-equipped to differentiate between normal and abnormal events. In this work, we formalize abnormal event detection as a one-versus-rest binary classification problem. Our contribution is two-fold. First, we introduce an unsupervised feature learning framework based on object-centric convolutional auto-encoders to encode both motion and appearance information. Second, we propose a supervised classification approach based on clustering the training samples into normality clusters. A one-versus-rest abnormal event classifier is then employed to separate each normality cluster from the rest. For the purpose of training the classifier, the other clusters act as dummy anomalies. During inference, an object is labeled as abnormal if the highest classification score assigned by the one-versus-rest classifiers is negative. Comprehensive experiments are performed on four benchmarks: Avenue, ShanghaiTech, UCSD and UMN. Our approach provides superior results on all four data sets. On the large-scale ShanghaiTech data set, our method provides an absolute gain of 8.4% in terms of frame-level AUC compared to the state-of-the-art method [Sultani et al., CVPR 2018].
https://arxiv.org/abs/1812.04960
Generating diverse yet specific data is the goal of the generative adversarial network (GAN), but it suffers from the problem of mode collapse. We introduce the concept of normalized diversity which forced the model to preserve the normalized pairwise distance between the sparse samples from a latent parametric distribution and their corresponding high-dimensional outputs. The normalized diversification aims to unfold the manifold of unknown topology and non-uniform distribution, which leads to safe interpolation between valid latent variables. By alternating the maximization over the pairwise distance and updating the total distance (normalizer), we encourage the model to actively explore in the high-dimensional output space. We demonstrate that by combining the normalized diversity loss and the adversarial loss, we generate diverse data without suffering from mode collapsing. Experimental results show that our method achieves consistent improvement on unsupervised image generation, conditional image generation and hand pose estimation over strong baselines.
http://arxiv.org/abs/1904.03608
Approaches to goal-directed behaviour including online planning and opportunistic planning tackle a change in the environment by generating alternative goals to avoid failures or seize opportunities. However, current approaches only address unanticipated changes related to objects or object types already defined in the planning task that is being solved. This article describes a domain-independent approach that advances the state of the art by extending the knowledge of a planning task with relevant objects of new types. The approach draws upon the use of ontologies, semantic measures, and ontology alignment to accommodate newly acquired data that trigger the formulation of goal opportunities inducing a better-valued plan.
http://arxiv.org/abs/1904.03606
Precise character segmentation is the only solution towards higher Optical Character Recognition (OCR) accuracy. In cursive script, overlapped characters are serious issue in the process of character segmentations as characters are deprived from their discriminative parts using conventional linear segmentation strategy. Hence, non-linear segmentation is an utmost need to avoid loss of characters parts and to enhance character/script recognition accuracy. This paper presents an improved approach for non-linear segmentation of the overlapped characters in handwritten roman script. The proposed technique is composed of a sequence of heuristic rules based on geometrical features of characters to locate possible non-linear character boundaries in a cursive script word. However, to enhance efficiency, heuristic approach is integrated with trained ensemble neural network validation strategy for verification of character boundaries. Accordingly, correct boundaries are retained and incorrect are removed based on ensemble neural networks vote. Finally, based on verified valid segmentation points, characters are segmented non-linearly. For fair comparison CEDAR benchmark database is experimented. The experimental results are much better than conventional linear character segmentation techniques reported in the state of art. Ensemble neural network play vital role to enhance character segmentation accuracy as compared to individual neural networks.
http://arxiv.org/abs/1904.12592
The cursive nature of multilingual characters segmentation and recognition of Arabic, Persian, Urdu languages have attracted researchers from academia and industry. However, despite several decades of research, still multilingual characters classification accuracy is not up to the mark. This paper presents an automated approach for multilingual characters segmentation and recognition. The proposed methodology explores character based on their geometric features. However, due to uncertainty and without dictionary support few characters are over-divided. To expand the productivity of the proposed methodology a BPN is prepared with countless division focuses for cursive multilingual characters. Prepared BPN separates off base portioned indicates effectively with rapid upgrade character acknowledgment precision. For reasonable examination, only benchmark dataset is utilized.
http://arxiv.org/abs/1904.08760
The Multi-target Challenge aims to assess how well current speech technology is able to determine whether or not a recorded utterance was spoken by one of a large number of blacklisted speakers. It is a form of multi-target speaker detection based on real-world telephone conversations. Data recordings are generated from call center customer-agent conversations. The task is to measure how accurately one can detect 1) whether a test recording is spoken by a blacklisted speaker, and 2) which specific blacklisted speaker was talking. This paper outlines the challenge and provides its baselines, results, and discussions.
http://arxiv.org/abs/1904.04240
In this paper, we propose VoiceID loss, a novel loss function for training a speech enhancement model to improve the robustness of speaker verification. In contrast to the commonly used loss functions for speech enhancement such as the L2 loss, the VoiceID loss is based on the feedback from a speaker verification model to generate a ratio mask. The generated ratio mask is multiplied pointwise with the original spectrogram to filter out unnecessary components for speaker verification. In the experiments, we observed that the enhancement network, after training with the VoiceID loss, is able to ignore a substantial amount of time-frequency bins, such as those dominated by noise, for verification. The resulting model consistently improves the speaker verification system on both clean and noisy conditions.
http://arxiv.org/abs/1904.03601
We address the problem of video representation learning without human-annotated labels. While previous efforts address the problem by designing novel self-supervised tasks using video data, the learned features are merely on a frame-by-frame basis, which are not applicable to many video analytic tasks where spatio-temporal features are prevailing. In this paper we propose a novel self-supervised approach to learn spatio-temporal features for video representation. Inspired by the success of two-stream approaches in video classification, we propose to learn visual features by regressing both motion and appearance statistics along spatial and temporal dimensions, given only the input video data. Specifically, we extract statistical concepts (fast-motion region and the corresponding dominant direction, spatio-temporal color diversity, dominant color, etc.) from simple patterns in both spatial and temporal domains. Unlike prior puzzles that are even hard for humans to solve, the proposed approach is consistent with human inherent visual habits and therefore easy to answer. We conduct extensive experiments with C3D to validate the effectiveness of our proposed approach. The experiments show that our approach can significantly improve the performance of C3D when applied to video classification tasks. Code is available at https://github.com/laura-wang/video_repres_mas.
http://arxiv.org/abs/1904.03597
Fine-tuning neural networks is widely used to transfer valuable knowledge from high-resource to low-resource domains. In a standard fine-tuning scheme, source and target problems are trained using the same architecture. Although capable of adapting to new domains, pre-trained units struggle with learning uncommon target-specific patterns. In this paper, we propose to augment the target-network with normalised, weighted and randomly initialised units that beget a better adaptation while maintaining the valuable source knowledge. Our experiments on POS tagging of social media texts (Tweets domain) demonstrate that our method achieves state-of-the-art performances on 3 commonly used datasets.
http://arxiv.org/abs/1904.03595
The adaptive moment estimation algorithm Adam (Kingma and Ba, ICLR 2015) is a popular optimizer in the training of deep neural networks. However, Reddi et al. (ICLR 2018) have recently shown that the convergence proof of Adam is problematic and proposed a variant of Adam called AMSGrad as a fix. In this paper, we show that the convergence proof of AMSGrad is also problematic, and we present various fixes for it, which include a new version of AMSGrad.
http://arxiv.org/abs/1904.03590
Computer Vision applications often require a textual grounding module with precision, interpretability, and resilience to counterfactual inputs/queries. To achieve high grounding precision, current textual grounding methods heavily rely on large-scale training data with manual annotations at the pixel level. Such annotations are expensive to obtain and thus severely narrow the model’s scope of real-world applications. Moreover, most of these methods sacrifice interpretability, generalizability, and they neglect the importance of being resilient to counterfactual inputs. To address these issues, we propose a visual grounding system which is 1) end-to-end trainable in a weakly supervised fashion with only image-level annotations, and 2) counterfactually resilient owing to the modular design. Specifically, we decompose textual descriptions into three levels: entity, semantic attribute, color information, and perform compositional grounding progressively. We validate our model through a series of experiments and demonstrate its improvement over the state-of-the-art methods. In particular, our model’s performance not only surpasses other weakly/un-supervised methods and even approaches the strongly supervised ones, but also is interpretable for decision making and performs much better in face of counterfactual classes than all the others.
http://arxiv.org/abs/1904.03589
The task of multi-label image recognition is to predict a set of object labels that present in an image. As objects normally co-occur in an image, it is desirable to model the label dependencies to improve the recognition performance. To capture and explore such important dependencies, we propose a multi-label classification model based on Graph Convolutional Network (GCN). The model builds a directed graph over the object labels, where each node (label) is represented by word embeddings of a label, and GCN is learned to map this label graph into a set of inter-dependent object classifiers. These classifiers are applied to the image descriptors extracted by another sub-net, enabling the whole network to be end-to-end trainable. Furthermore, we propose a novel re-weighted scheme to create an effective label correlation matrix to guide information propagation among the nodes in GCN. Experiments on two multi-label image recognition datasets show that our approach obviously outperforms other existing state-of-the-art methods. In addition, visualization analyses reveal that the classifiers learned by our model maintain meaningful semantic topology.
http://arxiv.org/abs/1904.03582
The recognition ability of human beings is developed in a progressive way. Usually, children learn to discriminate various objects from coarse to fine-grained with limited supervision. Inspired by this learning process, we propose a simple yet effective model for the Few-Shot Fine-Grained (FSFG) recognition, which tries to tackle the challenging fine-grained recognition task using meta-learning. The proposed method, named Pairwise Alignment Bilinear Network (PABN), is an end-to-end deep neural network. Unlike traditional deep bilinear networks for fine-grained classification, which adopt the self-bilinear pooling to capture the subtle features of images, the proposed model uses a novel pairwise bilinear pooling to compare the nuanced differences between base images and query images for learning a deep distance metric. In order to match base image features with query image features, we design feature alignment losses before the proposed pairwise bilinear pooling. Experiment results on four fine-grained classification datasets and one generic few-shot dataset demonstrate that the proposed model outperforms both the state-ofthe-art few-shot fine-grained and general few-shot methods.
http://arxiv.org/abs/1904.03580
This paper presents a novel adaptively connected neural network (ACNet) to improve the traditional convolutional neural networks (CNNs) {in} two aspects. First, ACNet employs a flexible way to switch global and local inference in processing the internal feature representations by adaptively determining the connection status among the feature nodes (e.g., pixels of the feature maps) \footnote{In a computer vision domain, a node refers to a pixel of a feature map{, while} in {the} graph domain, a node denotes a graph node.}. We can show that existing CNNs, the classical multilayer perceptron (MLP), and the recently proposed non-local network (NLN) \cite{nonlocalnn17} are all special cases of ACNet. Second, ACNet is also capable of handling non-Euclidean data. Extensive experimental analyses on {a variety of benchmarks (i.e.,} ImageNet-1k classification, COCO 2017 detection and segmentation, CUHK03 person re-identification, CIFAR analysis, and Cora document categorization) demonstrate that {ACNet} cannot only achieve state-of-the-art performance but also overcome the limitation of the conventional MLP and CNN \footnote{Corresponding author: Liang Lin (linliang@ieee.org)}. The code is available at \url{https://github.com/wanggrun/Adaptively-Connected-Neural-Networks}.
http://arxiv.org/abs/1904.03579
Decoding speaker’s intent is a crucial part of spoken language understanding (SLU). The presence of noise or errors in the text transcriptions, in real life scenarios make the task more challenging. In this paper, we address the spoken language intent detection under noisy conditions imposed by automatic speech recognition (ASR) systems. We propose to employ confusion2vec word feature representation to compensate for the errors made by ASR and to increase the robustness of the SLU system. The confusion2vec, motivated from human speech production and perception, models acoustic relationships between words in addition to the semantic and syntactic relations of words in human language. We hypothesize that ASR often makes errors relating to acoustically similar words, and the confusion2vec with inherent model of acoustic relationships between words is able to compensate for the errors. We demonstrate through experiments on the ATIS benchmark dataset, the robustness of the proposed model to achieve state-of-the-art results under noisy ASR conditions. Our system reduces classification error rate (CER) by 20.84% and improves robustness by 37.48% (lower CER degradation) relative to the previous state-of-the-art going from clean to noisy transcripts. Improvements are also demonstrated when training the intent detection models on noisy transcripts.
http://arxiv.org/abs/1904.03576
Recently, with the advent of deep convolutional neural networks (DCNN), the improvements in visual saliency prediction research are impressive. One possible direction to approach the next improvement is to fully characterize the multi-scale saliency-influential factors with a computationally-friendly module in DCNN architectures. In this work, we proposed an end-to-end dilated inception network (DINet) for visual saliency prediction. It captures multi-scale contextual features effectively with very limited extra parameters. Instead of utilizing parallel standard convolutions with different kernel sizes as the existing inception module, our proposed dilated inception module (DIM) uses parallel dilated convolutions with different dilation rates which can significantly reduce the computation load while enriching the diversity of receptive fields in feature maps. Moreover, the performance of our saliency model is further improved by using a set of linear normalization-based probability distribution distance metrics as loss functions. As such, we can formulate saliency prediction as a probability distribution prediction task for global saliency inference instead of a typical pixel-wise regression problem. Experimental results on several challenging saliency benchmark datasets demonstrate that our DINet with proposed loss functions can achieve state-of-the-art performance with shorter inference time.
http://arxiv.org/abs/1904.03571
Eating is an essential activity of daily living (ADL) for staying healthy and living at home independently. Although numerous assistive devices have been introduced, many people with disabilities are still restricted from independent eating due to the devices’ physical or perceptual limitations. In this work, we introduce a new meal-assistance system using a general-purpose mobile manipulator, a Willow Garage PR2, which has the potential to serve as a versatile form of assistive technology. Our active feeding framework enables the robot to autonomously deliver food to the user’s mouth. In detail, our web-based user interface, visually-guided behaviors, and safety tools allow people with severe motor impairments to benefit from the robotic assistance. We evaluated our system with 10 able-bodied participants and 9 people with motor impairments. Both groups of participants successfully ate various foods using the system and reported high rates of success for the system’s autonomous behaviors in a laboratory environment. Then, we performed in-home evaluation with Henry Evans, a person with quadriplegia, at his house in California, USA. In general, Henry and the other people who operated the system reported that it was comfortable, safe, and easy-to-use. We discuss learned lessons and design insights through user evaluations.
http://arxiv.org/abs/1904.03568
In recent years, the image and video coding technologies have advanced by leaps and bounds. However, due to the popularization of image and video acquisition devices, the growth rate of image and video data is far beyond the improvement of the compression ratio. In particular, it has been widely recognized that there are increasing challenges of pursuing further coding performance improvement within the traditional hybrid coding framework. Deep convolution neural network (CNN) which makes the neural network resurge in recent years and has achieved great success in both artificial intelligent and signal processing fields, also provides a novel and promising solution for image and video compression. In this paper, we provide a systematic, comprehensive and up-to-date review of neural network based image and video compression techniques. The evolution and development of neural network based compression methodologies are introduced for images and video respectively. More specifically, the cutting-edge video coding techniques by leveraging deep learning and HEVC framework are presented and discussed, which promote the state-of-the-art video coding performance substantially. Moreover, the end-to-end image and video coding frameworks based on neural networks are also reviewed, revealing interesting explorations on next generation image and video coding frameworks/standards. The most significant research works on the image and video coding related topics using neural networks are highlighted, and future trends are also envisioned. In particular, the joint compression on semantic and visual information is tentatively explored to formulate high efficiency signal representation structure for both human vision and machine vision, which are the two dominant signal receptor in the age of artificial intelligence.
http://arxiv.org/abs/1904.03567
Image change detection (ICD) to detect changed objects in front of a vehicle with respect to a place-specific background model using an on-board monocular vision system is a fundamental problem in intelligent vehicle (IV). From the perspective of recent large-scale IV applications, it can be impractical in terms of space/time efficiency to train place-specific background models for every possible place. To address these issues, we introduce a new autoencoder (AE) based efficient ICD framework that combines the advantages of AE-based anomaly detection (AD) and AE-based image compression (IC). We propose a method that uses AE reconstruction errors as a single unified measure for training a minimal set of place-specific AEs and maintains detection accuracy. We introduce an efficient incremental recursive AE (rAE) training framework that recursively summarizes a large collection of background images into the AE set. The results of experiments on challenging cross-season ICD tasks validate the efficacy of the proposed approach.
http://arxiv.org/abs/1904.03555
We present a novel scalable framework for image change detection (ICD) from an on-board 3D imagery system. We argue that existing ICD systems are constrained by the time required to align a given query image with individual reference image coordinates. We utilize an invariant coordinate system (ICS) to replace the time-consuming image alignment with an offline pre-processing procedure. Our key contribution is an extension of the traditional image comparison-based ICD tasks to setups of the image retrieval (IR) task. We replace each component of the 3D ICD system, i.e., (1) image modeling, (2) image alignment, and (3) image differencing, with significantly efficient variants from the bag-of-words (BoW) IR paradigm. Further, we train a deep 3D feature extractor in an unsupervised manner using an unsupervised Siamese network and automatically collected training data. We conducted experiments on a challenging cross-season ICD task using a publicly available dataset and thereby validate the efficacy of the proposed approach.
http://arxiv.org/abs/1904.03552
Most of the current state-of-the-art frameworks for cross-season visual place recognition (CS-VPR) focus on domain adaptation (DA) to a single specific season. From the viewpoint of long-term CS-VPR, such frameworks do not scale well to sequential multiple domains (e.g., spring - summer - autumn - winter - … ). The goal of this study is to develop a novel long-term ensemble learning (LEL) framework that allows for a constant cost retraining in long-term sequential-multi-domain CS-VPR (SMD-VPR), which only requires the memorization of a small constant number of deep convolutional neural networks (CNNs) and can retrain the CNN ensemble of every season at a small constant time/space cost. We frame our task as the multi-teacher multi-student knowledge distillation (MTMS-KD), which recursively compresses all the previous season’s knowledge into a current CNN ensemble. We further address the issue of teacher-student-assignment (TSA) to achieve a good generalization/specialization tradeoff. Experimental results on SMD-VPR tasks validate the efficacy of the proposed approach.
http://arxiv.org/abs/1904.03551
Acoustic scenes are rich and redundant in their content. In this work, we present a spatio-temporal attention pooling layer coupled with a convolutional recurrent neural network to learn from patterns that are discriminative while suppressing those that are irrelevant for acoustic scene classification. The convolutional layers in this network learn invariant features from time-frequency input. The bidirectional recurrent layers are then able to encode the temporal dynamics of the resulting convolutional features. Afterwards, a two-dimensional attention mask is formed via the outer product of the spatial and temporal attention vectors learned from two designated attention layers to weigh and pool the recurrent output into a final feature vector for classification. The network is trained with between-class examples generated from between-class data augmentation. Experiments demonstrate that the proposed method not only outperforms a strong convolutional neural network baseline but also sets new state-of-the-art performance on the LITIS Rouen dataset.
http://arxiv.org/abs/1904.03543
This paper presents a methodology for early detection of audio events from audio streams. Early detection is the ability to infer an ongoing event during its initial stage. The proposed system consists of a novel inference step coupled with dual parallel tailored-loss deep neural networks (DNNs). The DNNs share a similar architecture except for their loss functions, i.e. weighted loss and multitask loss, which are designed to efficiently cope with issues common to audio event detection. The inference step is newly introduced to make use of the network outputs for recognizing ongoing events. The monotonicity of the detection function is required for reliable early detection, and will also be proved. Experiments on the ITC-Irst database show that the proposed system achieves state-of-the-art detection performance. Furthermore, even partial events are sufficient to achieve good performance similar to that obtained when an entire event is observed, enabling early event detection.
http://arxiv.org/abs/1712.02116
There are few digital tools to help designers create game mechanics. A general language to express game mechanics is necessary for rapid game design iteration. The first iteration of a mechanics-focused language, together with its interfacing tool, are introduced in this paper. The language is restricted to two-dimensional, turn-based, tile-based, deterministic, complete-information games. The tool is compared to the existing alternatives for game mechanics prototyping and shown to be capable of succinctly implementing a range of well-known game mechanics.
http://arxiv.org/abs/1904.03540
Despite the advancement of supervised image recognition algorithms, their dependence on the availability of labeled data and the rapid expansion of image categories raise the significant challenge of zero-shot learning. Zero-shot learning (ZSL) aims to transfer knowledge from labeled classes into unlabeled classes to reduce human labeling effort. In this paper, we propose a novel progressive ensemble network model with multiple projected label embeddings to address zero-shot image recognition. The ensemble network is built by learning multiple image classification functions with a shared feature extraction network but different label embedding representations, which enhance the diversity of the classifiers and facilitate information transfer to unlabeled classes. A progressive training framework is then deployed to gradually label the most confident images in each unlabeled class with predicted pseudo-labels and update the ensemble network with the training data augmented by the pseudo-labels. The proposed model performs training on both labeled and unlabeled data. It can naturally bridge the domain shift problem in visual appearances and be extended to the generalized zero-shot learning scenario. We conduct experiments on multiple ZSL datasets and the empirical results demonstrate the efficacy of the proposed model.
http://arxiv.org/abs/1805.07473
In relation extraction with distant supervision, noisy labels make it difficult to train quality models. Previous neural models addressed this problem using an attention mechanism that attends to sentences that are likely to express the relations. We improve such models by combining the distant supervision data with an additional directly-supervised data, which we use as supervision for the attention weights. We find that joint training on both types of supervision leads to a better model because it improves the model’s ability to identify noisy sentences. In addition, we find that sigmoidal attention weights with max pooling achieves better performance over the commonly used weighted average attention in this setup. Our proposed method achieves a new state-of-the-art result on the widely used FB-NYT dataset.
http://arxiv.org/abs/1810.12956
Deep reinforcement learning (deep RL) has achieved superior performance in complex sequential tasks by using a deep neural network as its function approximator and by learning directly from raw images. A drawback of using raw images is that deep RL must learn the state feature representation from the raw images in addition to learning a policy. As a result, deep RL can require a prohibitively large amount of training time and data to reach reasonable performance, making it difficult to use deep RL in real-world applications, especially when data is expensive. In this work, we speed up training by addressing half of what deep RL is trying to solve — learning features. Our approach is to learn some of the important features by pre-training deep RL network’s hidden layers via supervised learning using a small set of human demonstrations. We empirically evaluate our approach using deep Q-network (DQN) and asynchronous advantage actor-critic (A3C) algorithms on the Atari 2600 games of Pong, Freeway, and Beamrider. Our results show that: 1) pre-training with human demonstrations in a supervised learning manner is better at discovering features relative to pre-training naively in DQN, and 2) initializing a deep RL network with a pre-trained model provides a significant improvement in training time even when pre-training from a small number of human demonstrations.
http://arxiv.org/abs/1709.04083
Backtracking line-search is an old yet powerful strategy for finding better step size to be used in proximal gradient algorithms. The main principle is to locally find a simple convex upper bound of the objective function, which in turn controls the step size that is used. In case of inertial proximal gradient algorithms, the situation becomes much more difficult and usually leads to very restrictive rules on the extrapolation parameter. In this paper, we show that the extrapolation parameter can be controlled by locally finding also a simple concave lower bound of the objective function. This gives rise to a double convex-concave backtracking procedure which allows for an adaptive and optimal choice of both the step size and extrapolation parameters. We apply this procedure to the class of inertial Bregman proximal gradient methods, and prove that any sequence generated converges globally to critical points of the function at hand. Numerical experiments on a number of challenging non-convex problems in image processing and machine learning were conducted and show the power of combining inertial step and double backtracking strategy in achieving improved performances.
http://arxiv.org/abs/1904.03537
We introduce Bayesian least-squares policy iteration (BLSPI), an off-policy, model-free, policy iteration algorithm that uses the Bayesian least-squares temporal-difference (BLSTD) learning algorithm to evaluate policies. An online variant of BLSPI has been also proposed, called randomised BLSPI (RBLSPI), that improves its policy based on an incomplete policy evaluation step. In online setting, the exploration-exploitation dilemma should be addressed as we try to discover the optimal policy by using samples collected by ourselves. RBLSPI exploits the advantage of BLSTD to quantify our uncertainty about the value function. Inspired by Thompson sampling, RBLSPI first samples a value function from a posterior distribution over value functions, and then selects actions based on the sampled value function. The effectiveness and the exploration abilities of RBLSPI are demonstrated experimentally in several environments.
http://arxiv.org/abs/1904.03535
Image segmentation is an important task in many medical applications. Methods based on convolutional neural networks attain state-of-the-art accuracy; however, they typically rely on supervised training with large labeled datasets. Labeling medical images requires significant expertise and time, and typical hand-tuned approaches for data augmentation fail to capture the complex variations in such images. We present an automated data augmentation method for synthesizing labeled medical images. We demonstrate our method on the task of segmenting magnetic resonance imaging (MRI) brain scans. Our method requires only a single segmented scan, and leverages other unlabeled scans in a semi-supervised approach. We learn a model of transformations from the images, and use the model along with the labeled example to synthesize additional labeled examples. Each transformation is comprised of a spatial deformation field and an intensity change, enabling the synthesis of complex effects such as variations in anatomy and image acquisition procedures. We show that training a supervised segmenter with these new examples provides significant improvements over state-of-the-art methods for one-shot biomedical image segmentation. Our code is available at https://github.com/xamyzhao/brainstorm.
http://arxiv.org/abs/1902.09383
The use of distances based on optimal transportation has recently shown promise for discrimination of power spectra. In particular, spectral estimation methods based on l1 regularization as well as covariance based methods can be shown to be robust with respect to such distances. These transportation distances provide a geometric framework where geodesics corresponds to smooth transition of spectral mass, and have been useful for tracking. In this paper, we investigate the use of these distances for automatic target recognition. We study the use of the Monge-Kantorovich distance compared to the standard l2 distance for classifying civilian vehicles based on SAR images. We use a version of the Monge-Kantorovich distance that applies also for the case where the spectra may have different total mass, and we formulate the optimization problem as a minimum flow problem that can be computed using efficient algorithms.
http://arxiv.org/abs/1904.03534
Decision-making under uncertainty is a crucial ability for autonomous systems. In its most general form, this problem can be formulated as a Partially Observable Markov Decision Process (POMDP). The solution policy of a POMDP can be implicitly encoded as a value function. In partially observable settings, the value function is typically learned via forward simulation of the system evolution. Focusing on accurate and long-range risk assessment, we propose a novel method, where the value function is learned in different phases via a bi-directional search in belief space. A backward value learning process provides a long-range and risk-aware base policy. A forward value learning process ensures local optimality and updates the policy via forward simulations. We consider a class of scalable and continuous-space rover navigation problems (RNP) to assess the safety, scalability, and optimality of the proposed algorithm. The results demonstrate the capabilities of the proposed algorithm in evaluating long-range risk/safety of the planner while addressing continuous problems with long planning horizons.
http://arxiv.org/abs/1902.05698
Model quantization is a widely used technique to compress and accelerate deep neural network (DNN) inference. Emergent DNN hardware accelerators begin to support mixed precision (1-8 bits) to further improve the computation efficiency, which raises a great challenge to find the optimal bitwidth for each layer: it requires domain experts to explore the vast design space trading off among accuracy, latency, energy, and model size, which is both time-consuming and sub-optimal. Conventional quantization algorithm ignores the different hardware architectures and quantizes all the layers in a uniform way. In this paper, we introduce the Hardware-Aware Automated Quantization (HAQ) framework which leverages the reinforcement learning to automatically determine the quantization policy, and we take the hardware accelerator’s feedback in the design loop. Rather than relying on proxy signals such as FLOPs and model size, we employ a hardware simulator to generate direct feedback signals (latency and energy) to the RL agent. Compared with conventional methods, our framework is fully automated and can specialize the quantization policy for different neural network architectures and hardware architectures. Our framework effectively reduced the latency by 1.4-1.95x and the energy consumption by 1.9x with negligible loss of accuracy compared with the fixed bitwidth (8 bits) quantization. Our framework reveals that the optimal policies on different hardware architectures (i.e., edge and cloud architectures) under different resource constraints (i.e., latency, energy and model size) are drastically different. We interpreted the implication of different quantization policies, which offer insights for both neural network architecture design and hardware architecture design.
http://arxiv.org/abs/1811.08886
3D Morphable Models (3DMMs) are statistical models that represent facial texture and shape variations using a set of linear bases and more particular Principal Component Analysis (PCA). 3DMMs were used as statistical priors for reconstructing 3D faces from images by solving non-linear least square optimization problems. Recently, 3DMMs were used as generative models for training non-linear mappings (\ie, regressors) from image to the parameters of the models via Deep Convolutional Neural Networks (DCNNs). Nevertheless, all of the above methods use either fully connected layers or 2D convolutions on parametric unwrapped UV spaces leading to large networks with many parameters. In this paper, we present the first, to the best of our knowledge, non-linear 3DMMs by learning joint texture and shape auto-encoders using direct mesh convolutions. We demonstrate how these auto-encoders can be used to train very light-weight models that perform Coloured Mesh Decoding (CMD) in-the-wild at a speed of over 2500 FPS.
http://arxiv.org/abs/1904.03525
This paper introduces Taco-VC, a novel architecture for voice conversion (VC) based on the Tacotron synthesizer, which is a sequence-to-sequence with attention model. Most current prosody preserving VC systems suffer from target similarity and quality issues in the converted speech. To address these problems, we first recover initial prosody preserving speech using a Phonetic Posteriorgrams (PPGs) based Tacotron synthesizer. Then, we enhance the quality of the converted speech using a novel speech-enhancement network, which is based on a combination of phoneme recognition and Tacotron networks. The final converted speech is generated by a Wavenet vocoder conditioned on Mel Spectrograms. Given the advantages of a single speaker Tacotron and Wavenet, we show how to adapt them to other speakers with limited training data. We evaluate our solution on the VCC 2018 SPOKE task. Using public mid-size datasets, our method outperforms the baseline and achieves competitive results
http://arxiv.org/abs/1904.03522
Procedural text, which describes entities and their interactions as they undergo some process, depicts entities in a uniquely nuanced way. First, each entity may have some observable discrete attributes, such as its state or location; modeling these involves imposing global structure and enforcing consistency. Second, an entity may have properties which are not made explicit but can be effectively induced and tracked by neural networks. In this paper, we propose a structured neural architecture that reflects this dual nature of entity evolution. The model tracks each entity recurrently, updating its hidden continuous representation at each step to contain relevant state information. The global discrete state structure is explicitly modeled with a neural CRF over the changing hidden representation of the entity. This CRF can explicitly capture constraints on entity states over time, enforcing that, for example, an entity cannot move to a location after it is destroyed. We evaluate the performance of our proposed model on QA tasks over process paragraphs in the ProPara dataset and find that our model achieves state-of-the-art results.
http://arxiv.org/abs/1904.03518
This paper presents a normalization mechanism called Instance-Level Meta Normalization (ILM~Norm) to address a learning-to-normalize problem. ILM~Norm learns to predict the normalization parameters via both the feature feed-forward and the gradient back-propagation paths. ILM~Norm provides a meta normalization mechanism and has several good properties. It can be easily plugged into existing instance-level normalization schemes such as Instance Normalization, Layer Normalization, or Group Normalization. ILM~Norm normalizes each instance individually and therefore maintains high performance even when small mini-batch is used. The experimental results show that ILM~Norm well adapts to different network architectures and tasks, and it consistently improves the performance of the original models. The code is available at url{https://github.com/Gasoonjia/ILM-Norm.
http://arxiv.org/abs/1904.03516
Efficient inference of Convolutional Neural Networks is a thriving topic recently. It is desirable to achieve the maximal test accuracy under given inference budget constraints when deploying a pre-trained model. Network pruning is a commonly used technique while it may produce irregular sparse models that can hardly gain actual speed-up. Group convolution is a promising pruning target due to its regular structure; however, incorporating such structure into the pruning procedure is challenging. It is because structural constraints are hard to describe and can make pruning intractable to solve. The need for configuring sparsity that maximises test accuracy also increases difficulty. This paper presents an efficient method to address this challenge. We formulate group convolution pruning as finding the optimal channel permutation to impose structural constraints and solve it efficiently by heuristics. We also apply local search to exploring the sparsity configuration that maximises test accuracy. Compared to prior work, results show that our method produces competitive group convolution models for various tasks within a shorter pruning period and enables rapid model sparsity exploration subject to inference budget constraints
http://arxiv.org/abs/1811.09341
Understanding procedural text requires tracking entities, actions and effects as the narrative unfolds. We focus on the challenging real-world problem of action-graph extraction from material science papers, where language is highly specialized and data annotation is expensive and scarce. We propose a novel approach, Text2Quest, where procedural text is interpreted as instructions for an interactive game. A learning agent completes the game by executing the procedure correctly in a text-based simulated lab environment. The framework can complement existing approaches and enables richer forms of learning compared to static texts. We discuss potential limitations and advantages of the approach, and release a prototype proof-of-concept, hoping to encourage research in this direction.
http://arxiv.org/abs/1811.04319
In this paper, we describe our submission to SemEval-2019 Task 4 on Hyperpartisan News Detection. Our system relies on a variety of engineered features originally used to detect propaganda. This is based on the assumption that biased messages are propagandistic in the sense that they promote a particular political cause or viewpoint. We trained a logistic regression model with features ranging from simple bag-of-words to vocabulary richness and text readability features. Our system achieved 72.9% accuracy on the test data that is annotated manually and 60.8% on the test data that is annotated with distant supervision. Additional experiments showed that significant performance improvements can be achieved with better feature pre-processing.
http://arxiv.org/abs/1904.03513
This paper describes a channel-selection approach for simplifying deep neural networks. Specifically, we propose a new type of generic network layer, called pruning layer, to seamlessly augment a given pre-trained model for compression. Each pruning layer, comprising $1 \times 1$ depth-wise kernels, is represented with a dual format: one is real-valued and the other is binary. The former enables a two-phase optimization process of network pruning to operate with an end-to-end differentiable network, and the latter yields the mask information for channel selection. Our method progressively performs the pruning task layer-wise, and achieves channel selection according to a sparsity criterion to favor pruning more channels. We also develop a cost-aware mechanism to prevent the compression from sacrificing the expected network performance. Our results for compressing several benchmark deep networks on image classification and semantic segmentation are comparable to those by state-of-the-art.
http://arxiv.org/abs/1904.03508
3D data is a valuable asset the computer vision filed as it provides rich information about the full geometry of sensed objects and scenes. Recently, with the availability of both large 3D datasets and computational power, it is today possible to consider applying deep learning to learn specific tasks on 3D data such as segmentation, recognition and correspondence. Depending on the considered 3D data representation, different challenges may be foreseen in using existent deep learning architectures. In this work, we provide a comprehensive overview about various 3D data representations highlighting the difference between Euclidean and non-Euclidean ones. We also discuss how Deep Learning methods are applied on each representation, analyzing the challenges to overcome.
http://arxiv.org/abs/1808.01462
Pulmonary nodule detection plays an important role in lung cancer screening with low-dose computed tomography (CT) scans. Although promising performance has been achieved by deep learning based nodule detection methods, it remains challenging to build nodule detection networks with good generalization performance due to unbalanced positive and negative samples. In order to overcome this problem and further improve state-of-the-art region proposal network methods, we develop a novel deep 3D convolutional neural network with an Encoder-Decoder structure for pulmonary nodule detection. Particularly, we utilize a dynamically scaled cross entropy loss to reduce the false positive rate and compensate the significant data imbalance problem. We adopt the squeeze-and-excitation structure to learn effective image features and fully utilize channel inter-dependency. We have validated our method based on publicly available CT scans from LIDC/IDRI dataset and its subset LUNA16 with thinner slices. Ablation studies and experimental results have demonstrated that our method could outperform state-of-the-art nodule detection methods by a large margin, with an average FROC score of 86.2% on LUNA16, and an average FROC score of 77.3% on LIDC/IDRI dataset when trained on LUNA16 only.
http://arxiv.org/abs/1904.03501
Traditional 3D Convolutional Neural Networks (CNNs) are computationally expensive, memory intensive, prone to overfit, and most importantly, there is a need to improve their feature learning capabilities. To address these issues, we propose Rectified Local Phase Volume (ReLPV) block, an efficient alternative to the standard 3D convolutional layer. The ReLPV block extracts the phase in a 3D local neighborhood (e.g., 3x3x3) of each position of the input map to obtain the feature maps. The phase is extracted by computing 3D Short Term Fourier Transform (STFT) at multiple fixed low frequency points in the 3D local neighborhood of each position. These feature maps at different frequency points are then linearly combined after passing them through an activation function. The ReLPV block provides significant parameter savings of at least, 3^3 to 13^3 times compared to the standard 3D convolutional layer with the filter sizes 3x3x3 to 13x13x13, respectively. We show that the feature learning capabilities of the ReLPV block are significantly better than the standard 3D convolutional layer. Furthermore, it produces consistently better results across different 3D data representations. We achieve state-of-the-art accuracy on the volumetric ModelNet10 and ModelNet40 datasets while utilizing only 11% parameters of the current state-of-the-art. We also improve the state-of-the-art on the UCF-101 split-1 action recognition dataset by 5.68% (when trained from scratch) while using only 15% of the parameters of the state-of-the-art. The project webpage is available at https://sites.google.com/view/lp-3dcnn/home.
http://arxiv.org/abs/1904.03498
We present a new large-scale multilingual video description dataset, VATEX, which contains over 41,250 videos and 825,000 captions in both English and Chinese. Among the captions, there are over 206,000 English-Chinese parallel translation pairs. Compared to the widely-used MSR-VTT dataset, VATEX is multilingual, larger, linguistically complex, and more diverse in terms of both video and natural language descriptions. We also introduce two tasks for video-and-language research based on VATEX: (1) Multilingual Video Captioning, aimed at describing a video in various languages with a compact unified captioning model, and (2) Video-guided Machine Translation, to translate a source language description into the target language using the video information as additional spatiotemporal context. Extensive experiments on the VATEX dataset show that, first, the unified multilingual model can not only produce both English and Chinese descriptions for a video more efficiently, but also offer improved performance over the monolingual models. Furthermore, we demonstrate that the spatiotemporal video context can be effectively utilized to align source and target languages and thus assist machine translation. In the end, we discuss the potentials of using VATEX for other video-and-language research.
http://arxiv.org/abs/1904.03493
A good representation for arbitrarily complicated data should have the capability of semantic generation, clustering and reconstruction. Previous research has already achieved impressive performance on either one. This paper aims at learning a disentangled representation effective for all of them in an unsupervised way. To achieve all the three tasks together, we learn the forward and inverse mapping between data and representation on the basis of a symmetric adversarial process. In theory, we minimize the upper bound of the two conditional entropy loss between the latent variables and the observations together to achieve the cycle consistency. The newly proposed RepGAN is tested on MNIST, fashionMNIST, CelebA, and SVHN datasets to perform unsupervised classification, generation and reconstruction tasks. The result demonstrates that RepGAN is able to learn a useful and competitive representation. To the author’s knowledge, our work is the first one to achieve both a high unsupervised classification accuracy and low reconstruction error on MNIST. Codes are available at https://github.com/yzhouas/RepGAN-tensorflow.
http://arxiv.org/abs/1804.07353
Contrary to i-vectors, speaker embeddings such as x-vectors are incapable of leveraging unlabelled utterances, due to the classification loss over training speakers. In this paper, we explore an alternative training strategy to enable the use of unlabelled utterances in training. We propose to train speaker embedding extractors via reconstructing the frames of a target speech segment, given the inferred embedding of another speech segment of the same utterance. We do this by attaching to the standard speaker embedding extractor a decoder network, which we feed not merely with the speaker embedding, but also with the estimated phone sequence of the target frame sequence. The reconstruction loss can be used either as a single objective, or be combined with the standard speaker classification loss. In the latter case, it acts as a regularizer, encouraging generalizability to speakers unseen during training. In all cases, the proposed architectures are trained from scratch and in an end-to-end fashion. We demonstrate the benefits from the proposed approach on VoxCeleb and Speakers in the wild, and we report notable improvements over the baseline.
http://arxiv.org/abs/1904.03486