October 20, 2019

3413 words 17 mins read

Paper Group AWR 342

Autoregressive Quantile Networks for Generative Modeling. Guiding Deep Learning System Testing using Surprise Adequacy. Deep Randomized Ensembles for Metric Learning. MEAL: Multi-Model Ensemble via Adversarial Learning. Temporal Pattern Attention for Multivariate Time Series Forecasting. The Global Anchor Method for Quantifying Linguistic Shifts an …

Autoregressive Quantile Networks for Generative Modeling


Title	Autoregressive Quantile Networks for Generative Modeling
Authors	Georg Ostrovski, Will Dabney, Rémi Munos
Abstract	We introduce autoregressive implicit quantile networks (AIQN), a fundamentally different approach to generative modeling than those commonly used, that implicitly captures the distribution using quantile regression. AIQN is able to achieve superior perceptual quality and improvements in evaluation metrics, without incurring a loss of sample diversity. The method can be applied to many existing models and architectures. In this work we extend the PixelCNN model with AIQN and demonstrate results on CIFAR-10 and ImageNet using Inception score, FID, non-cherry-picked samples, and inpainting results. We consistently observe that AIQN yields a highly stable algorithm that improves perceptual quality while maintaining a highly diverse distribution.
Tasks
Published	2018-06-14
URL	http://arxiv.org/abs/1806.05575v1
PDF	http://arxiv.org/pdf/1806.05575v1.pdf
PWC	https://paperswithcode.com/paper/autoregressive-quantile-networks-for
Repo	https://github.com/SSS135/aiqn-vae
Framework	pytorch

Guiding Deep Learning System Testing using Surprise Adequacy


Title	Guiding Deep Learning System Testing using Surprise Adequacy
Authors	Jinhan Kim, Robert Feldt, Shin Yoo
Abstract	Deep Learning (DL) systems are rapidly being adopted in safety and security critical domains, urgently calling for ways to test their correctness and robustness. Testing of DL systems has traditionally relied on manual collection and labelling of data. Recently, a number of coverage criteria based on neuron activation values have been proposed. These criteria essentially count the number of neurons whose activation during the execution of a DL system satisfied certain properties, such as being above predefined thresholds. However, existing coverage criteria are not sufficiently fine grained to capture subtle behaviours exhibited by DL systems. Moreover, evaluations have focused on showing correlation between adversarial examples and proposed criteria rather than evaluating and guiding their use for actual testing of DL systems. We propose a novel test adequacy criterion for testing of DL systems, called Surprise Adequacy for Deep Learning Systems (SADL), which is based on the behaviour of DL systems with respect to their training data. We measure the surprise of an input as the difference in DL system’s behaviour between the input and the training data (i.e., what was learnt during training), and subsequently develop this as an adequacy criterion: a good test input should be sufficiently but not overtly surprising compared to training data. Empirical evaluation using a range of DL systems from simple image classifiers to autonomous driving car platforms shows that systematic sampling of inputs based on their surprise can improve classification accuracy of DL systems against adversarial examples by up to 77.5% via retraining.
Tasks	Autonomous Driving
Published	2018-08-25
URL	http://arxiv.org/abs/1808.08444v1
PDF	http://arxiv.org/pdf/1808.08444v1.pdf
PWC	https://paperswithcode.com/paper/guiding-deep-learning-system-testing-using
Repo	https://github.com/lcckkkhaha/Suprise-Adequacy-Implementation
Framework	none

Deep Randomized Ensembles for Metric Learning


Title	Deep Randomized Ensembles for Metric Learning
Authors	Hong Xuan, Richard Souvenir, Robert Pless
Abstract	Learning embedding functions, which map semantically related inputs to nearby locations in a feature space supports a variety of classification and information retrieval tasks. In this work, we propose a novel, generalizable and fast method to define a family of embedding functions that can be used as an ensemble to give improved results. Each embedding function is learned by randomly bagging the training labels into small subsets. We show experimentally that these embedding ensembles create effective embedding functions. The ensemble output defines a metric space that improves state of the art performance for image retrieval on CUB-200-2011, Cars-196, In-Shop Clothes Retrieval and VehicleID.
Tasks	Image Retrieval, Information Retrieval, Metric Learning
Published	2018-08-13
URL	http://arxiv.org/abs/1808.04469v2
PDF	http://arxiv.org/pdf/1808.04469v2.pdf
PWC	https://paperswithcode.com/paper/deep-randomized-ensembles-for-metric-learning
Repo	https://github.com/littleredxh/DREML
Framework	pytorch

MEAL: Multi-Model Ensemble via Adversarial Learning


Title	MEAL: Multi-Model Ensemble via Adversarial Learning
Authors	Zhiqiang Shen, Zhankui He, Xiangyang Xue
Abstract	Often the best performing deep neural models are ensembles of multiple base-level networks. Unfortunately, the space required to store these many networks, and the time required to execute them at test-time, prohibits their use in applications where test sets are large (e.g., ImageNet). In this paper, we present a method for compressing large, complex trained ensembles into a single network, where knowledge from a variety of trained deep neural networks (DNNs) is distilled and transferred to a single DNN. In order to distill diverse knowledge from different trained (teacher) models, we propose to use adversarial-based learning strategy where we define a block-wise training loss to guide and optimize the predefined student network to recover the knowledge in teacher models, and to promote the discriminator network to distinguish teacher vs. student features simultaneously. The proposed ensemble method (MEAL) of transferring distilled knowledge with adversarial learning exhibits three important advantages: (1) the student network that learns the distilled knowledge with discriminators is optimized better than the original model; (2) fast inference is realized by a single forward pass, while the performance is even better than traditional ensembles from multi-original models; (3) the student network can learn the distilled knowledge from a teacher model that has arbitrary structures. Extensive experiments on CIFAR-10/100, SVHN and ImageNet datasets demonstrate the effectiveness of our MEAL method. On ImageNet, our ResNet-50 based MEAL achieves top-1/5 21.79%/5.99% val error, which outperforms the original model by 2.06%/1.14%. Code and models are available at: https://github.com/AaronHeee/MEAL
Tasks
Published	2018-12-06
URL	https://arxiv.org/abs/1812.02425v2
PDF	https://arxiv.org/pdf/1812.02425v2.pdf
PWC	https://paperswithcode.com/paper/meal-multi-model-ensemble-via-adversarial
Repo	https://github.com/AaronHeee/MEAL
Framework	pytorch

Temporal Pattern Attention for Multivariate Time Series Forecasting


Title	Temporal Pattern Attention for Multivariate Time Series Forecasting
Authors	Shun-Yao Shih, Fan-Keng Sun, Hung-yi Lee
Abstract	Forecasting multivariate time series data, such as prediction of electricity consumption, solar power production, and polyphonic piano pieces, has numerous valuable applications. However, complex and non-linear interdependencies between time steps and series complicate the task. To obtain accurate prediction, it is crucial to model long-term dependency in time series data, which can be achieved to some good extent by recurrent neural network (RNN) with attention mechanism. Typical attention mechanism reviews the information at each previous time step and selects the relevant information to help generate the outputs, but it fails to capture the temporal patterns across multiple time steps. In this paper, we propose to use a set of filters to extract time-invariant temporal patterns, which is similar to transforming time series data into its “frequency domain”. Then we proposed a novel attention mechanism to select relevant time series, and use its “frequency domain” information for forecasting. We applied the proposed model on several real-world tasks and achieved state-of-the-art performance in all of them with only one exception.
Tasks	Multivariate Time Series Forecasting, Time Series, Time Series Forecasting
Published	2018-09-12
URL	https://arxiv.org/abs/1809.04206v3
PDF	https://arxiv.org/pdf/1809.04206v3.pdf
PWC	https://paperswithcode.com/paper/temporal-pattern-attention-for-multivariate
Repo	https://github.com/abinashsinha330/Air-Pollution-Forecasting-using-Recurrent-Neural-Networks
Framework	tf

The Global Anchor Method for Quantifying Linguistic Shifts and Domain Adaptation


Title	The Global Anchor Method for Quantifying Linguistic Shifts and Domain Adaptation
Authors	Zi Yin, Vin Sachidananda, Balaji Prabhakar
Abstract	Language is dynamic, constantly evolving and adapting with respect to time, domain or topic. The adaptability of language is an active research area, where researchers discover social, cultural and domain-specific changes in language using distributional tools such as word embeddings. In this paper, we introduce the global anchor method for detecting corpus-level language shifts. We show both theoretically and empirically that the global anchor method is equivalent to the alignment method, a widely-used method for comparing word embeddings, in terms of detecting corpus-level language shifts. Despite their equivalence in terms of detection abilities, we demonstrate that the global anchor method is superior in terms of applicability as it can compare embeddings of different dimensionalities. Furthermore, the global anchor method has implementation and parallelization advantages. We show that the global anchor method reveals fine structures in the evolution of language and domain adaptation. When combined with the graph Laplacian technique, the global anchor method recovers the evolution trajectory and domain clustering of disparate text corpora.
Tasks	Domain Adaptation, Word Embeddings
Published	2018-12-12
URL	http://arxiv.org/abs/1812.10382v1
PDF	http://arxiv.org/pdf/1812.10382v1.pdf
PWC	https://paperswithcode.com/paper/the-global-anchor-method-for-quantifying
Repo	https://github.com/ziyin-dl/global-anchor-method
Framework	none

Adapting multi-armed bandits policies to contextual bandits scenarios


Title	Adapting multi-armed bandits policies to contextual bandits scenarios
Authors	David Cortes
Abstract	This work explores adaptations of successful multi-armed bandits policies to the online contextual bandits scenario with binary rewards using binary classification algorithms such as logistic regression as black-box oracles. Some of these adaptations are achieved through bootstrapping or approximate bootstrapping, while others rely on other forms of randomness, resulting in more scalable approaches than previous works, and the ability to work with any type of classification algorithm. In particular, the Adaptive-Greedy algorithm shows a lot of promise, in many cases achieving better performance than upper confidence bound and Thompson sampling strategies, at the expense of more hyperparameters to tune.
Tasks	Multi-Armed Bandits
Published	2018-11-11
URL	https://arxiv.org/abs/1811.04383v2
PDF	https://arxiv.org/pdf/1811.04383v2.pdf
PWC	https://paperswithcode.com/paper/adapting-multi-armed-bandits-policies-to
Repo	https://github.com/david-cortes/contextualbandits
Framework	none

Capsules for Object Segmentation


Title	Capsules for Object Segmentation
Authors	Rodney LaLonde, Ulas Bagci
Abstract	Convolutional neural networks (CNNs) have shown remarkable results over the last several years for a wide range of computer vision tasks. A new architecture recently introduced by Sabour et al., referred to as a capsule networks with dynamic routing, has shown great initial results for digit recognition and small image classification. The success of capsule networks lies in their ability to preserve more information about the input by replacing max-pooling layers with convolutional strides and dynamic routing, allowing for preservation of part-whole relationships in the data. This preservation of the input is demonstrated by reconstructing the input from the output capsule vectors. Our work expands the use of capsule networks to the task of object segmentation for the first time in the literature. We extend the idea of convolutional capsules with locally-connected routing and propose the concept of deconvolutional capsules. Further, we extend the masked reconstruction to reconstruct the positive input class. The proposed convolutional-deconvolutional capsule network, called SegCaps, shows strong results for the task of object segmentation with substantial decrease in parameter space. As an example application, we applied the proposed SegCaps to segment pathological lungs from low dose CT scans and compared its accuracy and efficiency with other U-Net-based architectures. SegCaps is able to handle large image sizes (512 x 512) as opposed to baseline capsules (typically less than 32 x 32). The proposed SegCaps reduced the number of parameters of U-Net architecture by 95.4% while still providing a better segmentation accuracy.
Tasks	Semantic Segmentation
Published	2018-04-11
URL	http://arxiv.org/abs/1804.04241v1
PDF	http://arxiv.org/pdf/1804.04241v1.pdf
PWC	https://paperswithcode.com/paper/capsules-for-object-segmentation
Repo	https://github.com/Cheng-Lin-Li/SegCaps
Framework	tf

Fully Decentralized Multi-Agent Reinforcement Learning with Networked Agents


Title	Fully Decentralized Multi-Agent Reinforcement Learning with Networked Agents
Authors	Kaiqing Zhang, Zhuoran Yang, Han Liu, Tong Zhang, Tamer Başar
Abstract	We consider the problem of \emph{fully decentralized} multi-agent reinforcement learning (MARL), where the agents are located at the nodes of a time-varying communication network. Specifically, we assume that the reward functions of the agents might correspond to different tasks, and are only known to the corresponding agent. Moreover, each agent makes individual decisions based on both the information observed locally and the messages received from its neighbors over the network. Within this setting, the collective goal of the agents is to maximize the globally averaged return over the network through exchanging information with their neighbors. To this end, we propose two decentralized actor-critic algorithms with function approximation, which are applicable to large-scale MARL problems where both the number of states and the number of agents are massively large. Under the decentralized structure, the actor step is performed individually by each agent with no need to infer the policies of others. For the critic step, we propose a consensus update via communication over the network. Our algorithms are fully incremental and can be implemented in an online fashion. Convergence analyses of the algorithms are provided when the value functions are approximated within the class of linear functions. Extensive simulation results with both linear and nonlinear function approximations are presented to validate the proposed algorithms. Our work appears to be the first study of fully decentralized MARL algorithms for networked agents with function approximation, with provable convergence guarantees.
Tasks	Multi-agent Reinforcement Learning
Published	2018-02-23
URL	http://arxiv.org/abs/1802.08757v2
PDF	http://arxiv.org/pdf/1802.08757v2.pdf
PWC	https://paperswithcode.com/paper/fully-decentralized-multi-agent-reinforcement
Repo	https://github.com/MUmarJaved/MultiAgent-Distributed-Reinforcement-Learning
Framework	tf

Deep Reinforcement Learning


Title	Deep Reinforcement Learning
Authors	Yuxi Li
Abstract	We discuss deep reinforcement learning in an overview style. We draw a big picture, filled with details. We discuss six core elements, six important mechanisms, and twelve applications, focusing on contemporary work, and in historical contexts. We start with background of artificial intelligence, machine learning, deep learning, and reinforcement learning (RL), with resources. Next we discuss RL core elements, including value function, policy, reward, model, exploration vs. exploitation, and representation. Then we discuss important mechanisms for RL, including attention and memory, unsupervised learning, hierarchical RL, multi-agent RL, relational RL, and learning to learn. After that, we discuss RL applications, including games, robotics, natural language processing (NLP), computer vision, finance, business management, healthcare, education, energy, transportation, computer systems, and, science, engineering, and art. Finally we summarize briefly, discuss challenges and opportunities, and close with an epilogue.
Tasks
Published	2018-10-15
URL	http://arxiv.org/abs/1810.06339v1
PDF	http://arxiv.org/pdf/1810.06339v1.pdf
PWC	https://paperswithcode.com/paper/deep-reinforcement-learning
Repo	https://github.com/google/dopamine
Framework	tf

Machine Teaching of Active Sequential Learners


Title	Machine Teaching of Active Sequential Learners
Authors	Tomi Peltola, Mustafa Mert Çelikok, Pedram Daee, Samuel Kaski
Abstract	Machine teaching addresses the problem of finding the best training data that can guide a learning algorithm to a target model with minimal effort. In conventional settings, a teacher provides data that are consistent with the true data distribution. However, for sequential learners which actively choose their queries, such as multi-armed bandits and active learners, the teacher can only provide responses to the learner’s queries, not design the full data. In this setting, consistent teachers can be sub-optimal for finite horizons. We formulate this sequential teaching problem, which current techniques in machine teaching do not address, as a Markov decision process, with the dynamics nesting a model of the learner and the actions being the teacher’s responses. Furthermore, we address the complementary problem of learning from a teacher that plans: to recognise the teaching intent of the responses, the learner is endowed with a model of the teacher. We test the formulation with multi-armed bandit learners in simulated experiments and a user study. The results show that learning is improved by (i) planning teaching and (ii) the learner having a model of the teacher. The approach gives tools to taking into account strategic (planning) behaviour of users of interactive intelligent systems, such as recommendation engines, by considering them as boundedly optimal teachers.
Tasks	Multi-Armed Bandits, Probabilistic Programming
Published	2018-09-08
URL	https://arxiv.org/abs/1809.02869v3
PDF	https://arxiv.org/pdf/1809.02869v3.pdf
PWC	https://paperswithcode.com/paper/modelling-users-theory-of-ais-mind-in
Repo	https://github.com/AaltoPML/machine-teaching-of-active-sequential-learners
Framework	pytorch

On-line Adaptative Curriculum Learning for GANs


Title	On-line Adaptative Curriculum Learning for GANs
Authors	Thang Doan, Joao Monteiro, Isabela Albuquerque, Bogdan Mazoure, Audrey Durand, Joelle Pineau, R Devon Hjelm
Abstract	Generative Adversarial Networks (GANs) can successfully approximate a probability distribution and produce realistic samples. However, open questions such as sufficient convergence conditions and mode collapse still persist. In this paper, we build on existing work in the area by proposing a novel framework for training the generator against an ensemble of discriminator networks, which can be seen as a one-student/multiple-teachers setting. We formalize this problem within the full-information adversarial bandit framework, where we evaluate the capability of an algorithm to select mixtures of discriminators for providing the generator with feedback during learning. To this end, we propose a reward function which reflects the progress made by the generator and dynamically update the mixture weights allocated to each discriminator. We also draw connections between our algorithm and stochastic optimization methods and then show that existing approaches using multiple discriminators in literature can be recovered from our framework. We argue that less expressive discriminators are smoother and have a general coarse grained view of the modes map, which enforces the generator to cover a wide portion of the data distribution support. On the other hand, highly expressive discriminators ensure samples quality. Finally, experimental results show that our approach improves samples quality and diversity over existing baselines by effectively learning a curriculum. These results also support the claim that weaker discriminators have higher entropy improving modes coverage. Keywords: multiple discriminators, curriculum learning, multiple resolutions discriminators, multi-armed bandits, generative adversarial networks, smooth discriminators, multi-discriminator gan training, multiple experts.
Tasks	Multi-Armed Bandits, Stochastic Optimization
Published	2018-07-31
URL	http://arxiv.org/abs/1808.00020v6
PDF	http://arxiv.org/pdf/1808.00020v6.pdf
PWC	https://paperswithcode.com/paper/on-line-adaptative-curriculum-learning-for
Repo	https://github.com/Byte7/Adaptative-Curriculum-GAN-keras
Framework	none

PAC-Bayes Control: Learning Policies that Provably Generalize to Novel Environments


Title	PAC-Bayes Control: Learning Policies that Provably Generalize to Novel Environments
Authors	Anirudha Majumdar, Alec Farid, Anoopkumar Sonar
Abstract	Our goal is to learn control policies for robots that provably generalize well to novel environments given a dataset of example environments. The key technical idea behind our approach is to leverage tools from generalization theory in machine learning by exploiting a precise analogy (which we present in the form of a reduction) between generalization of control policies to novel environments and generalization of hypotheses in the supervised learning setting. In particular, we utilize the Probably Approximately Correct (PAC)-Bayes framework, which allows us to obtain upper bounds that hold with high probability on the expected cost of (stochastic) control policies across novel environments. We propose policy learning algorithms that explicitly seek to minimize this upper bound. The corresponding optimization problem can be solved using convex optimization (Relative Entropy Programming in particular) in the setting where we are optimizing over a finite policy space. In the more general setting of continuously parameterized policies (e.g., neural network policies), we minimize this upper bound using stochastic gradient descent. We present simulated results of our approach applied to learning (1) reactive obstacle avoidance policies and (2) neural network-based grasping policies. We also present hardware results for the Parrot Swing drone navigating through different obstacle environments. Our examples demonstrate the potential of our approach to provide strong generalization guarantees for robotic systems with continuous state and action spaces, complicated (e.g., nonlinear) dynamics, rich sensory inputs (e.g., depth images), and neural network-based policies.
Tasks
Published	2018-06-11
URL	https://arxiv.org/abs/1806.04225v4
PDF	https://arxiv.org/pdf/1806.04225v4.pdf
PWC	https://paperswithcode.com/paper/pac-bayes-control-synthesizing-controllers
Repo	https://github.com/irom-lab/PAC-Bayes-Control
Framework	none

Graph R-CNN for Scene Graph Generation


Title	Graph R-CNN for Scene Graph Generation
Authors	Jianwei Yang, Jiasen Lu, Stefan Lee, Dhruv Batra, Devi Parikh
Abstract	We propose a novel scene graph generation model called Graph R-CNN, that is both effective and efficient at detecting objects and their relations in images. Our model contains a Relation Proposal Network (RePN) that efficiently deals with the quadratic number of potential relations between objects in an image. We also propose an attentional Graph Convolutional Network (aGCN) that effectively captures contextual information between objects and relations. Finally, we introduce a new evaluation metric that is more holistic and realistic than existing metrics. We report state-of-the-art performance on scene graph generation as evaluated using both existing and our proposed metrics.
Tasks	Graph Generation, Scene Graph Generation
Published	2018-08-01
URL	http://arxiv.org/abs/1808.00191v1
PDF	http://arxiv.org/pdf/1808.00191v1.pdf
PWC	https://paperswithcode.com/paper/graph-r-cnn-for-scene-graph-generation
Repo	https://github.com/jwyang/graph-rcnn.pytorch
Framework	pytorch

RGBiD-SLAM for Accurate Real-time Localisation and 3D Mapping


Title	RGBiD-SLAM for Accurate Real-time Localisation and 3D Mapping
Authors	Daniel Gutierrez-Gomez, Jose J. Guerrero
Abstract	In this paper we present a complete SLAM system for RGB-D cameras, namely RGB-iD SLAM. The presented approach is a dense direct SLAM method with the main characteristic of working with the depth maps in inverse depth parametrisation for the routines of dense alignment or keyframe fusion. The system consists in 2 CPU threads working in parallel, which share the use of the GPU for dense alignment and keyframe fusion routines. The first thread is a front-end operating at frame rate, which processes every incoming frame from the RGB-D sensor to compute the incremental odometry and integrate it in a keyframe which is changed periodically following a covisibility-based strategy. The second thread is a back-end which receives keyframes from the front-end. This thread is in charge of segmenting the keyframes based on their structure, describing them using Bags of Words, trying to find potential loop closures with previous keyframes, and in such case perform pose-graph optimisation for trajectory correction. In addition, our system allows is able to compute the odometry both with unregistered and registered depth maps, allowing to use customised calibrations of the RGB-D sensor. As a consequence in the paper we also propose a detailed calibration pipeline to compute customised calibrations for particular RGB-D cameras. The experiments with our approach in the TUM RGB-D benchmark datasets show results superior in accuracy to the state-of-the-art in many of the sequences. The code has been made available on-line for research purposes https://github.com/dangut/RGBiD-SLAM.
Tasks	Calibration
Published	2018-07-22
URL	http://arxiv.org/abs/1807.08271v1
PDF	http://arxiv.org/pdf/1807.08271v1.pdf
PWC	https://paperswithcode.com/paper/rgbid-slam-for-accurate-real-time
Repo	https://github.com/dangut/RGBiD-SLAM
Framework	none