July 30, 2019

2910 words 14 mins read

Paper Group AWR 60

TreeQN and ATreeC: Differentiable Tree-Structured Models for Deep Reinforcement Learning. Distributed Mapping with Privacy and Communication Constraints: Lightweight Algorithms and Object-based Models. Ensemble representation learning: an analysis of fitness and survival for wrapper-based genetic programming methods. Interactive Attention Networks …

TreeQN and ATreeC: Differentiable Tree-Structured Models for Deep Reinforcement Learning


Title	TreeQN and ATreeC: Differentiable Tree-Structured Models for Deep Reinforcement Learning
Authors	Gregory Farquhar, Tim Rocktäschel, Maximilian Igl, Shimon Whiteson
Abstract	Combining deep model-free reinforcement learning with on-line planning is a promising approach to building on the successes of deep RL. On-line planning with look-ahead trees has proven successful in environments where transition models are known a priori. However, in complex environments where transition models need to be learned from data, the deficiencies of learned models have limited their utility for planning. To address these challenges, we propose TreeQN, a differentiable, recursive, tree-structured model that serves as a drop-in replacement for any value function network in deep RL with discrete actions. TreeQN dynamically constructs a tree by recursively applying a transition model in a learned abstract state space and then aggregating predicted rewards and state-values using a tree backup to estimate Q-values. We also propose ATreeC, an actor-critic variant that augments TreeQN with a softmax layer to form a stochastic policy network. Both approaches are trained end-to-end, such that the learned model is optimised for its actual use in the tree. We show that TreeQN and ATreeC outperform n-step DQN and A2C on a box-pushing task, as well as n-step DQN and value prediction networks (Oh et al. 2017) on multiple Atari games. Furthermore, we present ablation studies that demonstrate the effect of different auxiliary losses on learning transition models.
Tasks	Atari Games
Published	2017-10-31
URL	http://arxiv.org/abs/1710.11417v2
PDF	http://arxiv.org/pdf/1710.11417v2.pdf
PWC	https://paperswithcode.com/paper/treeqn-and-atreec-differentiable-tree
Repo	https://github.com/oxwhirl/treeqn
Framework	pytorch

Distributed Mapping with Privacy and Communication Constraints: Lightweight Algorithms and Object-based Models


Title	Distributed Mapping with Privacy and Communication Constraints: Lightweight Algorithms and Object-based Models
Authors	Siddharth Choudhary, Luca Carlone, Carlos Nieto, John Rogers, Henrik I. Christensen, Frank Dellaert
Abstract	We consider the following problem: a team of robots is deployed in an unknown environment and it has to collaboratively build a map of the area without a reliable infrastructure for communication. The backbone for modern mapping techniques is pose graph optimization, which estimates the trajectory of the robots, from which the map can be easily built. The first contribution of this paper is a set of distributed algorithms for pose graph optimization: rather than sending all sensor data to a remote sensor fusion server, the robots exchange very partial and noisy information to reach an agreement on the pose graph configuration. Our approach can be considered as a distributed implementation of the two-stage approach of Carlone et al., where we use the Successive Over-Relaxation (SOR) and the Jacobi Over-Relaxation (JOR) as workhorses to split the computation among the robots. As a second contribution, we extend %and demonstrate the applicability of the proposed distributed algorithms to work with object-based map models. The use of object-based models avoids the exchange of raw sensor measurements (e.g., point clouds) further reducing the communication burden. Our third contribution is an extensive experimental evaluation of the proposed techniques, including tests in realistic Gazebo simulations and field experiments in a military test facility. Abundant experimental evidence suggests that one of the proposed algorithms (the Distributed Gauss-Seidel method or DGS) has excellent performance. The DGS requires minimal information exchange, has an anytime flavor, scales well to large teams, is robust to noise, and is easy to implement. Our field tests show that the combined use of our distributed algorithms and object-based models reduces the communication requirements by several orders of magnitude and enables distributed mapping with large teams of robots in real-world problems.
Tasks	Sensor Fusion
Published	2017-02-11
URL	http://arxiv.org/abs/1702.03435v1
PDF	http://arxiv.org/pdf/1702.03435v1.pdf
PWC	https://paperswithcode.com/paper/distributed-mapping-with-privacy-and
Repo	https://github.com/CogRob/distributed-mapper
Framework	none

Ensemble representation learning: an analysis of fitness and survival for wrapper-based genetic programming methods


Title	Ensemble representation learning: an analysis of fitness and survival for wrapper-based genetic programming methods
Authors	William La Cava, Jason H. Moore
Abstract	Recently we proposed a general, ensemble-based feature engineering wrapper (FEW) that was paired with a number of machine learning methods to solve regression problems. Here, we adapt FEW for supervised classification and perform a thorough analysis of fitness and survival methods within this framework. Our tests demonstrate that two fitness metrics, one introduced as an adaptation of the silhouette score, outperform the more commonly used Fisher criterion. We analyze survival methods and demonstrate that $\epsilon$-lexicase survival works best across our test problems, followed by random survival which outperforms both tournament and deterministic crowding. We conduct a benchmark comparison to several classification methods using a large set of problems and show that FEW can improve the best classifier performance in several cases. We show that FEW generates consistent, meaningful features for a biomedical problem with different ML pairings.
Tasks	Feature Engineering, Representation Learning
Published	2017-03-20
URL	http://arxiv.org/abs/1703.06934v3
PDF	http://arxiv.org/pdf/1703.06934v3.pdf
PWC	https://paperswithcode.com/paper/ensemble-representation-learning-an-analysis
Repo	https://github.com/lacava/few
Framework	none

Interactive Attention Networks for Aspect-Level Sentiment Classification


Title	Interactive Attention Networks for Aspect-Level Sentiment Classification
Authors	Dehong Ma, Sujian Li, Xiaodong Zhang, Houfeng Wang
Abstract	Aspect-level sentiment classification aims at identifying the sentiment polarity of specific target in its context. Previous approaches have realized the importance of targets in sentiment classification and developed various methods with the goal of precisely modeling their contexts via generating target-specific representations. However, these studies always ignore the separate modeling of targets. In this paper, we argue that both targets and contexts deserve special treatment and need to be learned their own representations via interactive learning. Then, we propose the interactive attention networks (IAN) to interactively learn attentions in the contexts and targets, and generate the representations for targets and contexts separately. With this design, the IAN model can well represent a target and its collocative context, which is helpful to sentiment classification. Experimental results on SemEval 2014 Datasets demonstrate the effectiveness of our model.
Tasks	Aspect-Based Sentiment Analysis
Published	2017-09-04
URL	http://arxiv.org/abs/1709.00893v1
PDF	http://arxiv.org/pdf/1709.00893v1.pdf
PWC	https://paperswithcode.com/paper/interactive-attention-networks-for-aspect
Repo	https://github.com/songyouwei/ABSA-PyTorch
Framework	pytorch

Regularizing and Optimizing LSTM Language Models


Title	Regularizing and Optimizing LSTM Language Models
Authors	Stephen Merity, Nitish Shirish Keskar, Richard Socher
Abstract	Recurrent neural networks (RNNs), such as long short-term memory networks (LSTMs), serve as a fundamental building block for many sequence learning tasks, including machine translation, language modeling, and question answering. In this paper, we consider the specific problem of word-level language modeling and investigate strategies for regularizing and optimizing LSTM-based models. We propose the weight-dropped LSTM which uses DropConnect on hidden-to-hidden weights as a form of recurrent regularization. Further, we introduce NT-ASGD, a variant of the averaged stochastic gradient method, wherein the averaging trigger is determined using a non-monotonic condition as opposed to being tuned by the user. Using these and other regularization strategies, we achieve state-of-the-art word level perplexities on two data sets: 57.3 on Penn Treebank and 65.8 on WikiText-2. In exploring the effectiveness of a neural cache in conjunction with our proposed model, we achieve an even lower state-of-the-art perplexity of 52.8 on Penn Treebank and 52.0 on WikiText-2.
Tasks	Language Modelling
Published	2017-08-07
URL	http://arxiv.org/abs/1708.02182v1
PDF	http://arxiv.org/pdf/1708.02182v1.pdf
PWC	https://paperswithcode.com/paper/regularizing-and-optimizing-lstm-language
Repo	https://github.com/prajjwal1/language-modelling
Framework	pytorch

Octree Generating Networks: Efficient Convolutional Architectures for High-resolution 3D Outputs


Title	Octree Generating Networks: Efficient Convolutional Architectures for High-resolution 3D Outputs
Authors	Maxim Tatarchenko, Alexey Dosovitskiy, Thomas Brox
Abstract	We present a deep convolutional decoder architecture that can generate volumetric 3D outputs in a compute- and memory-efficient manner by using an octree representation. The network learns to predict both the structure of the octree, and the occupancy values of individual cells. This makes it a particularly valuable technique for generating 3D shapes. In contrast to standard decoders acting on regular voxel grids, the architecture does not have cubic complexity. This allows representing much higher resolution outputs with a limited memory budget. We demonstrate this in several application domains, including 3D convolutional autoencoders, generation of objects and whole scenes from high-level representations, and shape from a single image.
Tasks	3D Reconstruction
Published	2017-03-28
URL	http://arxiv.org/abs/1703.09438v3
PDF	http://arxiv.org/pdf/1703.09438v3.pdf
PWC	https://paperswithcode.com/paper/octree-generating-networks-efficient
Repo	https://github.com/lmb-freiburg/ogn
Framework	none

Large-scale, Fast and Accurate Shot Boundary Detection through Spatio-temporal Convolutional Neural Networks


Title	Large-scale, Fast and Accurate Shot Boundary Detection through Spatio-temporal Convolutional Neural Networks
Authors	Ahmed Hassanien, Mohamed Elgharib, Ahmed Selim, Sung-Ho Bae, Mohamed Hefeeda, Wojciech Matusik
Abstract	Shot boundary detection (SBD) is an important pre-processing step for video manipulation. Here, each segment of frames is classified as either sharp, gradual or no transition. Current SBD techniques analyze hand-crafted features and attempt to optimize both detection accuracy and processing speed. However, the heavy computations of optical flow prevents this. To achieve this aim, we present an SBD technique based on spatio-temporal Convolutional Neural Networks (CNN). Since current datasets are not large enough to train an accurate SBD CNN, we present a new dataset containing more than 3.5 million frames of sharp and gradual transitions. The transitions are generated synthetically using image compositing models. Our dataset contain additional 70,000 frames of important hard-negative no transitions. We perform the largest evaluation to date for one SBD algorithm, on real and synthetic data, containing more than 4.85 million frames. In comparison to the state of the art, we outperform dissolve gradual detection, generate competitive performance for sharp detections and produce significant improvement in wipes. In addition, we are up to 11 times faster than the state of the art.
Tasks	Boundary Detection, Optical Flow Estimation
Published	2017-05-09
URL	http://arxiv.org/abs/1705.03281v2
PDF	http://arxiv.org/pdf/1705.03281v2.pdf
PWC	https://paperswithcode.com/paper/large-scale-fast-and-accurate-shot-boundary
Repo	https://github.com/Tangshitao/ClipShots
Framework	none

BAS: Beetle Antennae Search Algorithm for Optimization Problems


Title	BAS: Beetle Antennae Search Algorithm for Optimization Problems
Authors	Xiangyuan Jiang, Shuai Li
Abstract	Meta-heuristic algorithms have become very popular because of powerful performance on the optimization problem. A new algorithm called beetle antennae search algorithm (BAS) is proposed in the paper inspired by the searching behavior of longhorn beetles. The BAS algorithm imitates the function of antennae and the random walking mechanism of beetles in nature, and then two main steps of detecting and searching are implemented. Finally, the algorithm is benchmarked on 2 well-known test functions, in which the numerical results validate the efficacy of the proposed BAS algorithm.
Tasks
Published	2017-10-30
URL	http://arxiv.org/abs/1710.10724v1
PDF	http://arxiv.org/pdf/1710.10724v1.pdf
PWC	https://paperswithcode.com/paper/bas-beetle-antennae-search-algorithm-for
Repo	https://github.com/jywang2016/rBAS
Framework	none

Examining Cooperation in Visual Dialog Models


Title	Examining Cooperation in Visual Dialog Models
Authors	Mircea Mironenco, Dana Kianfar, Ke Tran, Evangelos Kanoulas, Efstratios Gavves
Abstract	In this work we propose a blackbox intervention method for visual dialog models, with the aim of assessing the contribution of individual linguistic or visual components. Concretely, we conduct structured or randomized interventions that aim to impair an individual component of the model, and observe changes in task performance. We reproduce a state-of-the-art visual dialog model and demonstrate that our methodology yields surprising insights, namely that both dialog and image information have minimal contributions to task performance. The intervention method presented here can be applied as a sanity check for the strength and robustness of each component in visual dialog systems.
Tasks	Visual Dialog
Published	2017-12-04
URL	http://arxiv.org/abs/1712.01329v1
PDF	http://arxiv.org/pdf/1712.01329v1.pdf
PWC	https://paperswithcode.com/paper/examining-cooperation-in-visual-dialog-models
Repo	https://github.com/danakianfar/Examining-Cooperation-in-VDM
Framework	pytorch

Retinal Vessel Segmentation in Fundoscopic Images with Generative Adversarial Networks


Title	Retinal Vessel Segmentation in Fundoscopic Images with Generative Adversarial Networks
Authors	Jaemin Son, Sang Jun Park, Kyu-Hwan Jung
Abstract	Retinal vessel segmentation is an indispensable step for automatic detection of retinal diseases with fundoscopic images. Though many approaches have been proposed, existing methods tend to miss fine vessels or allow false positives at terminal branches. Let alone under-segmentation, over-segmentation is also problematic when quantitative studies need to measure the precise width of vessels. In this paper, we present a method that generates the precise map of retinal vessels using generative adversarial training. Our methods achieve dice coefficient of 0.829 on DRIVE dataset and 0.834 on STARE dataset which is the state-of-the-art performance on both datasets.
Tasks	Retinal Vessel Segmentation
Published	2017-06-28
URL	http://arxiv.org/abs/1706.09318v1
PDF	http://arxiv.org/pdf/1706.09318v1.pdf
PWC	https://paperswithcode.com/paper/retinal-vessel-segmentation-in-fundoscopic
Repo	https://github.com/ChengBinJin/V-GAN-tensorflow
Framework	tf

Large Batch Training of Convolutional Networks


Title	Large Batch Training of Convolutional Networks
Authors	Yang You, Igor Gitman, Boris Ginsburg
Abstract	A common way to speed up training of large convolutional networks is to add computational units. Training is then performed using data-parallel synchronous Stochastic Gradient Descent (SGD) with mini-batch divided between computational units. With an increase in the number of nodes, the batch size grows. But training with large batch size often results in the lower model accuracy. We argue that the current recipe for large batch training (linear learning rate scaling with warm-up) is not general enough and training may diverge. To overcome this optimization difficulties we propose a new training algorithm based on Layer-wise Adaptive Rate Scaling (LARS). Using LARS, we scaled Alexnet up to a batch size of 8K, and Resnet-50 to a batch size of 32K without loss in accuracy.
Tasks
Published	2017-08-13
URL	http://arxiv.org/abs/1708.03888v3
PDF	http://arxiv.org/pdf/1708.03888v3.pdf
PWC	https://paperswithcode.com/paper/large-batch-training-of-convolutional
Repo	https://github.com/jxbz/fromage
Framework	pytorch

Two-sample Statistics Based on Anisotropic Kernels


Title	Two-sample Statistics Based on Anisotropic Kernels
Authors	Xiuyuan Cheng, Alexander Cloninger, Ronald R. Coifman
Abstract	The paper introduces a new kernel-based Maximum Mean Discrepancy (MMD) statistic for measuring the distance between two distributions given finitely-many multivariate samples. When the distributions are locally low-dimensional, the proposed test can be made more powerful to distinguish certain alternatives by incorporating local covariance matrices and constructing an anisotropic kernel. The kernel matrix is asymmetric; it computes the affinity between $n$ data points and a set of $n_R$ reference points, where $n_R$ can be drastically smaller than $n$. While the proposed statistic can be viewed as a special class of Reproducing Kernel Hilbert Space MMD, the consistency of the test is proved, under mild assumptions of the kernel, as long as $\p-q\ \sqrt{n} \to \infty $, and a finite-sample lower bound of the testing power is obtained. Applications to flow cytometry and diffusion MRI datasets are demonstrated, which motivate the proposed approach to compare distributions.
Tasks
Published	2017-09-14
URL	http://arxiv.org/abs/1709.05006v3
PDF	http://arxiv.org/pdf/1709.05006v3.pdf
PWC	https://paperswithcode.com/paper/two-sample-statistics-based-on-anisotropic
Repo	https://github.com/AClon42/two-sample-anisotropic
Framework	none

R-FCN-3000 at 30fps: Decoupling Detection and Classification


Title	R-FCN-3000 at 30fps: Decoupling Detection and Classification
Authors	Bharat Singh, Hengduo Li, Abhishek Sharma, Larry S. Davis
Abstract	We present R-FCN-3000, a large-scale real-time object detector in which objectness detection and classification are decoupled. To obtain the detection score for an RoI, we multiply the objectness score with the fine-grained classification score. Our approach is a modification of the R-FCN architecture in which position-sensitive filters are shared across different object classes for performing localization. For fine-grained classification, these position-sensitive filters are not needed. R-FCN-3000 obtains an mAP of 34.9% on the ImageNet detection dataset and outperforms YOLO-9000 by 18% while processing 30 images per second. We also show that the objectness learned by R-FCN-3000 generalizes to novel classes and the performance increases with the number of training object classes - supporting the hypothesis that it is possible to learn a universal objectness detector. Code will be made available.
Tasks
Published	2017-12-05
URL	http://arxiv.org/abs/1712.01802v1
PDF	http://arxiv.org/pdf/1712.01802v1.pdf
PWC	https://paperswithcode.com/paper/r-fcn-3000-at-30fps-decoupling-detection-and
Repo	https://github.com/starimpact/arm_SNIPER
Framework	tf

SLAM with Objects using a Nonparametric Pose Graph


Title	SLAM with Objects using a Nonparametric Pose Graph
Authors	Beipeng Mu, Shih-Yuan Liu, Liam Paull, John Leonard, Jonathan How
Abstract	Mapping and self-localization in unknown environments are fundamental capabilities in many robotic applications. These tasks typically involve the identification of objects as unique features or landmarks, which requires the objects both to be detected and then assigned a unique identifier that can be maintained when viewed from different perspectives and in different images. The \textit{data association} and \textit{simultaneous localization and mapping} (SLAM) problems are, individually, well-studied in the literature. But these two problems are inherently tightly coupled, and that has not been well-addressed. Without accurate SLAM, possible data associations are combinatorial and become intractable easily. Without accurate data association, the error of SLAM algorithms diverge easily. This paper proposes a novel nonparametric pose graph that models data association and SLAM in a single framework. An algorithm is further introduced to alternate between inferring data association and performing SLAM. Experimental results show that our approach has the new capability of associating object detections and localizing objects at the same time, leading to significantly better performance on both the data association and SLAM problems than achieved by considering only one and ignoring imperfections in the other.
Tasks	Simultaneous Localization and Mapping
Published	2017-04-19
URL	http://arxiv.org/abs/1704.05959v1
PDF	http://arxiv.org/pdf/1704.05959v1.pdf
PWC	https://paperswithcode.com/paper/slam-with-objects-using-a-nonparametric-pose
Repo	https://github.com/tiev-tongji/quadric_slam
Framework	none

Learning Robust Rewards with Adversarial Inverse Reinforcement Learning


Title	Learning Robust Rewards with Adversarial Inverse Reinforcement Learning
Authors	Justin Fu, Katie Luo, Sergey Levine
Abstract	Reinforcement learning provides a powerful and general framework for decision making and control, but its application in practice is often hindered by the need for extensive feature and reward engineering. Deep reinforcement learning methods can remove the need for explicit engineering of policy or value features, but still require a manually specified reward function. Inverse reinforcement learning holds the promise of automatic reward acquisition, but has proven exceptionally difficult to apply to large, high-dimensional problems with unknown dynamics. In this work, we propose adverserial inverse reinforcement learning (AIRL), a practical and scalable inverse reinforcement learning algorithm based on an adversarial reward learning formulation. We demonstrate that AIRL is able to recover reward functions that are robust to changes in dynamics, enabling us to learn policies even under significant variation in the environment seen during training. Our experiments show that AIRL greatly outperforms prior methods in these transfer settings.
Tasks	Decision Making
Published	2017-10-30
URL	http://arxiv.org/abs/1710.11248v2
PDF	http://arxiv.org/pdf/1710.11248v2.pdf
PWC	https://paperswithcode.com/paper/learning-robust-rewards-with-adversarial
Repo	https://github.com/HumanCompatibleAI/airl
Framework	none