July 29, 2019

3146 words 15 mins read

Paper Group AWR 126

Natural Language Does Not Emerge ‘Naturally’ in Multi-Agent Dialog. Variational Approaches for Auto-Encoding Generative Adversarial Networks. Transform-Invariant Non-Parametric Clustering of Covariance Matrices and its Application to Unsupervised Joint Segmentation and Action Discovery. Safe Policy Improvement with Baseline Bootstrapping. Guided De …

Natural Language Does Not Emerge ‘Naturally’ in Multi-Agent Dialog


Title	Natural Language Does Not Emerge ‘Naturally’ in Multi-Agent Dialog
Authors	Satwik Kottur, José M. F. Moura, Stefan Lee, Dhruv Batra
Abstract	A number of recent works have proposed techniques for end-to-end learning of communication protocols among cooperative multi-agent populations, and have simultaneously found the emergence of grounded human-interpretable language in the protocols developed by the agents, all learned without any human supervision! In this paper, using a Task and Tell reference game between two agents as a testbed, we present a sequence of ‘negative’ results culminating in a ‘positive’ one – showing that while most agent-invented languages are effective (i.e. achieve near-perfect task rewards), they are decidedly not interpretable or compositional. In essence, we find that natural language does not emerge ‘naturally’, despite the semblance of ease of natural-language-emergence that one may gather from recent literature. We discuss how it is possible to coax the invented languages to become more and more human-like and compositional by increasing restrictions on how two agents may communicate.
Tasks
Published	2017-06-26
URL	http://arxiv.org/abs/1706.08502v3
PDF	http://arxiv.org/pdf/1706.08502v3.pdf
PWC	https://paperswithcode.com/paper/natural-language-does-not-emerge-naturally-in
Repo	https://github.com/kdexd/lang-emerge-parlai
Framework	pytorch

Variational Approaches for Auto-Encoding Generative Adversarial Networks


Title	Variational Approaches for Auto-Encoding Generative Adversarial Networks
Authors	Mihaela Rosca, Balaji Lakshminarayanan, David Warde-Farley, Shakir Mohamed
Abstract	Auto-encoding generative adversarial networks (GANs) combine the standard GAN algorithm, which discriminates between real and model-generated data, with a reconstruction loss given by an auto-encoder. Such models aim to prevent mode collapse in the learned generative model by ensuring that it is grounded in all the available training data. In this paper, we develop a principle upon which auto-encoders can be combined with generative adversarial networks by exploiting the hierarchical structure of the generative model. The underlying principle shows that variational inference can be used a basic tool for learning, but with the in- tractable likelihood replaced by a synthetic likelihood, and the unknown posterior distribution replaced by an implicit distribution; both synthetic likelihoods and implicit posterior distributions can be learned using discriminators. This allows us to develop a natural fusion of variational auto-encoders and generative adversarial networks, combining the best of both these methods. We describe a unified objective for optimization, discuss the constraints needed to guide learning, connect to the wide range of existing work, and use a battery of tests to systematically and quantitatively assess the performance of our method.
Tasks
Published	2017-06-15
URL	http://arxiv.org/abs/1706.04987v2
PDF	http://arxiv.org/pdf/1706.04987v2.pdf
PWC	https://paperswithcode.com/paper/variational-approaches-for-auto-encoding
Repo	https://github.com/PrateekMunjal/Alpha_GAN
Framework	tf

Transform-Invariant Non-Parametric Clustering of Covariance Matrices and its Application to Unsupervised Joint Segmentation and Action Discovery


Title	Transform-Invariant Non-Parametric Clustering of Covariance Matrices and its Application to Unsupervised Joint Segmentation and Action Discovery
Authors	Nadia Figueroa, Aude Billard
Abstract	In this work, we tackle the problem of transform-invariant unsupervised learning in the space of Covariance matrices and applications thereof. We begin by introducing the Spectral Polytope Covariance Matrix (SPCM) Similarity function; a similarity function for Covariance matrices, invariant to any type of transformation. We then derive the SPCM-CRP mixture model, a transform-invariant non-parametric clustering approach for Covariance matrices that leverages the proposed similarity function, spectral embedding and the distance-dependent Chinese Restaurant Process (dd-CRP) (Blei and Frazier, 2011). The scalability and applicability of these two contributions is extensively validated on real-world Covariance matrix datasets from diverse research fields. Finally, we couple the SPCM-CRP mixture model with the Bayesian non-parametric Indian Buffet Process (IBP) - Hidden Markov Model (HMM) (Fox et al., 2009), to jointly segment and discover transform-invariant action primitives from complex sequential data. Resulting in a topic-modeling inspired hierarchical model for unsupervised time-series data analysis which we call ICSC-HMM (IBP Coupled SPCM-CRP Hidden Markov Model). The ICSC-HMM is validated on kinesthetic demonstrations of uni-manual and bi-manual cooking tasks; achieving unsupervised human-level decomposition of complex sequential tasks.
Tasks	Time Series
Published	2017-10-27
URL	http://arxiv.org/abs/1710.10060v1
PDF	http://arxiv.org/pdf/1710.10060v1.pdf
PWC	https://paperswithcode.com/paper/transform-invariant-non-parametric-clustering
Repo	https://github.com/nbfigueroa/SPCM-CRP
Framework	none

Safe Policy Improvement with Baseline Bootstrapping


Title	Safe Policy Improvement with Baseline Bootstrapping
Authors	Romain Laroche, Paul Trichelair, Rémi Tachet des Combes
Abstract	This paper considers Safe Policy Improvement (SPI) in Batch Reinforcement Learning (Batch RL): from a fixed dataset and without direct access to the true environment, train a policy that is guaranteed to perform at least as well as the baseline policy used to collect the data. Our approach, called SPI with Baseline Bootstrapping (SPIBB), is inspired by the knows-what-it-knows paradigm: it bootstraps the trained policy with the baseline when the uncertainty is high. Our first algorithm, $\Pi_b$-SPIBB, comes with SPI theoretical guarantees. We also implement a variant, $\Pi_{\leq b}$-SPIBB, that is even more efficient in practice. We apply our algorithms to a motivational stochastic gridworld domain and further demonstrate on randomly generated MDPs the superiority of SPIBB with respect to existing algorithms, not only in safety but also in mean performance. Finally, we implement a model-free version of SPIBB and show its benefits on a navigation task with deep RL implementation called SPIBB-DQN, which is, to the best of our knowledge, the first RL algorithm relying on a neural network representation able to train efficiently and reliably from batch data, without any interaction with the environment.
Tasks
Published	2017-12-19
URL	https://arxiv.org/abs/1712.06924v5
PDF	https://arxiv.org/pdf/1712.06924v5.pdf
PWC	https://paperswithcode.com/paper/safe-policy-improvement-with-baseline
Repo	https://github.com/RomainLaroche/SPIBB
Framework	none

Guided Deep Reinforcement Learning for Swarm Systems


Title	Guided Deep Reinforcement Learning for Swarm Systems
Authors	Maximilian Hüttenrauch, Adrian Šošić, Gerhard Neumann
Abstract	In this paper, we investigate how to learn to control a group of cooperative agents with limited sensing capabilities such as robot swarms. The agents have only very basic sensor capabilities, yet in a group they can accomplish sophisticated tasks, such as distributed assembly or search and rescue tasks. Learning a policy for a group of agents is difficult due to distributed partial observability of the state. Here, we follow a guided approach where a critic has central access to the global state during learning, which simplifies the policy evaluation problem from a reinforcement learning point of view. For example, we can get the positions of all robots of the swarm using a camera image of a scene. This camera image is only available to the critic and not to the control policies of the robots. We follow an actor-critic approach, where the actors base their decisions only on locally sensed information. In contrast, the critic is learned based on the true global state. Our algorithm uses deep reinforcement learning to approximate both the Q-function and the policy. The performance of the algorithm is evaluated on two tasks with simple simulated 2D agents: 1) finding and maintaining a certain distance to each others and 2) locating a target.
Tasks
Published	2017-09-18
URL	http://arxiv.org/abs/1709.06011v1
PDF	http://arxiv.org/pdf/1709.06011v1.pdf
PWC	https://paperswithcode.com/paper/guided-deep-reinforcement-learning-for-swarm
Repo	https://github.com/nsrishankar/rl_swarm_papers
Framework	none

RaspiReader: Open Source Fingerprint Reader


Title	RaspiReader: Open Source Fingerprint Reader
Authors	Joshua J. Engelsma, Kai Cao, Anil K. Jain
Abstract	We open source an easy to assemble, spoof resistant, high resolution, optical fingerprint reader, called RaspiReader, using ubiquitous components. By using our open source STL files and software, RaspiReader can be built in under one hour for only US $175. As such, RaspiReader provides the fingerprint research community a seamless and simple method for quickly prototyping new ideas involving fingerprint reader hardware. In particular, we posit that this open source fingerprint reader will facilitate the exploration of novel fingerprint spoof detection techniques involving both hardware and software. We demonstrate one such spoof detection technique by specially customizing RaspiReader with two cameras for fingerprint image acquisition. One camera provides high contrast, frustrated total internal reflection (FTIR) fingerprint images, and the other outputs direct images of the finger in contact with the platen. Using both of these image streams, we extract complementary information which, when fused together and used for spoof detection, results in marked performance improvement over previous methods relying only on grayscale FTIR images provided by COTS optical readers. Finally, fingerprint matching experiments between images acquired from the FTIR output of RaspiReader and images acquired from a COTS reader verify the interoperability of the RaspiReader with existing COTS optical readers.
Tasks
Published	2017-12-26
URL	http://arxiv.org/abs/1712.09392v1
PDF	http://arxiv.org/pdf/1712.09392v1.pdf
PWC	https://paperswithcode.com/paper/raspireader-open-source-fingerprint-reader
Repo	https://github.com/engelsjo/RaspiReader
Framework	none

Functional Map of the World


Title	Functional Map of the World
Authors	Gordon Christie, Neil Fendley, James Wilson, Ryan Mukherjee
Abstract	We present a new dataset, Functional Map of the World (fMoW), which aims to inspire the development of machine learning models capable of predicting the functional purpose of buildings and land use from temporal sequences of satellite images and a rich set of metadata features. The metadata provided with each image enables reasoning about location, time, sun angles, physical sizes, and other features when making predictions about objects in the image. Our dataset consists of over 1 million images from over 200 countries. For each image, we provide at least one bounding box annotation containing one of 63 categories, including a “false detection” category. We present an analysis of the dataset along with baseline approaches that reason about metadata and temporal views. Our data, code, and pretrained models have been made publicly available.
Tasks
Published	2017-11-21
URL	http://arxiv.org/abs/1711.07846v3
PDF	http://arxiv.org/pdf/1711.07846v3.pdf
PWC	https://paperswithcode.com/paper/functional-map-of-the-world
Repo	https://github.com/fmow/baseline
Framework	tf

APE-GAN: Adversarial Perturbation Elimination with GAN


Title	APE-GAN: Adversarial Perturbation Elimination with GAN
Authors	Shiwei Shen, Guoqing Jin, Ke Gao, Yongdong Zhang
Abstract	Although neural networks could achieve state-of-the-art performance while recongnizing images, they often suffer a tremendous defeat from adversarial examples–inputs generated by utilizing imperceptible but intentional perturbation to clean samples from the datasets. How to defense against adversarial examples is an important problem which is well worth researching. So far, very few methods have provided a significant defense to adversarial examples. In this paper, a novel idea is proposed and an effective framework based Generative Adversarial Nets named APE-GAN is implemented to defense against the adversarial examples. The experimental results on three benchmark datasets including MNIST, CIFAR10 and ImageNet indicate that APE-GAN is effective to resist adversarial examples generated from five attacks.
Tasks
Published	2017-07-18
URL	http://arxiv.org/abs/1707.05474v3
PDF	http://arxiv.org/pdf/1707.05474v3.pdf
PWC	https://paperswithcode.com/paper/ape-gan-adversarial-perturbation-elimination
Repo	https://github.com/carlini/APE-GAN
Framework	tf

Behavior Trees in Robotics and AI: An Introduction


Title	Behavior Trees in Robotics and AI: An Introduction
Authors	Michele Colledanchise, Petter Ögren
Abstract	A Behavior Tree (BT) is a way to structure the switching between different tasks in an autonomous agent, such as a robot or a virtual entity in a computer game. BTs are a very efficient way of creating complex systems that are both modular and reactive. These properties are crucial in many applications, which has led to the spread of BT from computer game programming to many branches of AI and Robotics. In this book, we will first give an introduction to BTs, then we describe how BTs relate to, and in many cases generalize, earlier switching structures. These ideas are then used as a foundation for a set of efficient and easy to use design principles. Properties such as safety, robustness, and efficiency are important for an autonomous system, and we describe a set of tools for formally analyzing these using a state space description of BTs. With the new analysis tools, we can formalize the descriptions of how BTs generalize earlier approaches. We also show the use of BTs in automated planning and machine learning. Finally, we describe an extended set of tools to capture the behavior of Stochastic BTs, where the outcomes of actions are described by probabilities. These tools enable the computation of both success probabilities and time to completion.
Tasks
Published	2017-08-31
URL	http://arxiv.org/abs/1709.00084v3
PDF	http://arxiv.org/pdf/1709.00084v3.pdf
PWC	https://paperswithcode.com/paper/behavior-trees-in-robotics-and-ai-an
Repo	https://github.com/miccol/ROS-Behavior-Tree
Framework	none

Automatically Generating Commit Messages from Diffs using Neural Machine Translation


Title	Automatically Generating Commit Messages from Diffs using Neural Machine Translation
Authors	Siyuan Jiang, Ameer Armaly, Collin McMillan
Abstract	Commit messages are a valuable resource in comprehension of software evolution, since they provide a record of changes such as feature additions and bug repairs. Unfortunately, programmers often neglect to write good commit messages. Different techniques have been proposed to help programmers by automatically writing these messages. These techniques are effective at describing what changed, but are often verbose and lack context for understanding the rationale behind a change. In contrast, humans write messages that are short and summarize the high level rationale. In this paper, we adapt Neural Machine Translation (NMT) to automatically “translate” diffs into commit messages. We trained an NMT algorithm using a corpus of diffs and human-written commit messages from the top 1k Github projects. We designed a filter to help ensure that we only trained the algorithm on higher-quality commit messages. Our evaluation uncovered a pattern in which the messages we generate tend to be either very high or very low quality. Therefore, we created a quality-assurance filter to detect cases in which we are unable to produce good messages, and return a warning instead.
Tasks	Machine Translation
Published	2017-08-30
URL	http://arxiv.org/abs/1708.09492v1
PDF	http://arxiv.org/pdf/1708.09492v1.pdf
PWC	https://paperswithcode.com/paper/automatically-generating-commit-messages-from
Repo	https://github.com/vladislavneon/autogenerating-commit-messages
Framework	none

STN-OCR: A single Neural Network for Text Detection and Text Recognition


Title	STN-OCR: A single Neural Network for Text Detection and Text Recognition
Authors	Christian Bartz, Haojin Yang, Christoph Meinel
Abstract	Detecting and recognizing text in natural scene images is a challenging, yet not completely solved task. In re- cent years several new systems that try to solve at least one of the two sub-tasks (text detection and text recognition) have been proposed. In this paper we present STN-OCR, a step towards semi-supervised neural networks for scene text recognition, that can be optimized end-to-end. In contrast to most existing works that consist of multiple deep neural networks and several pre-processing steps we propose to use a single deep neural network that learns to detect and recognize text from natural images in a semi-supervised way. STN-OCR is a network that integrates and jointly learns a spatial transformer network, that can learn to detect text regions in an image, and a text recognition network that takes the identified text regions and recognizes their textual content. We investigate how our model behaves on a range of different tasks (detection and recognition of characters, and lines of text). Experimental results on public benchmark datasets show the ability of our model to handle a variety of different tasks, without substantial changes in its overall network structure.
Tasks	Optical Character Recognition, Scene Text Detection, Scene Text Recognition
Published	2017-07-27
URL	http://arxiv.org/abs/1707.08831v1
PDF	http://arxiv.org/pdf/1707.08831v1.pdf
PWC	https://paperswithcode.com/paper/stn-ocr-a-single-neural-network-for-text
Repo	https://github.com/Narp99/FRI_2018
Framework	tf

Chinese Typography Transfer


Title	Chinese Typography Transfer
Authors	Jie Chang, Yujun Gu
Abstract	In this paper, we propose a new network architecture for Chinese typography transformation based on deep learning. The architecture consists of two sub-networks: (1)a fully convolutional network(FCN) aiming at transferring specified typography style to another in condition of preserving structure information; (2)an adversarial network aiming at generating more realistic strokes in some details. Unlike models proposed before 2012 relying on the complex segmentation of Chinese components or strokes, our model treats every Chinese character as an inseparable image, so pre-processing or post-preprocessing are abandoned. Besides, our model adopts end-to-end training without pre-trained used in other deep models. The experiments demonstrates that our model can synthesize realistic-looking target typography from any source typography both on printed style and handwriting style.
Tasks
Published	2017-07-16
URL	http://arxiv.org/abs/1707.04904v2
PDF	http://arxiv.org/pdf/1707.04904v2.pdf
PWC	https://paperswithcode.com/paper/chinese-typography-transfer
Repo	https://github.com/hanfeisun/Unsupervised-Typography-Transfer
Framework	tf

Fast Spectral Clustering Using Autoencoders and Landmarks


Title	Fast Spectral Clustering Using Autoencoders and Landmarks
Authors	Ershad Banijamali, Ali Ghodsi
Abstract	In this paper, we introduce an algorithm for performing spectral clustering efficiently. Spectral clustering is a powerful clustering algorithm that suffers from high computational complexity, due to eigen decomposition. In this work, we first build the adjacency matrix of the corresponding graph of the dataset. To build this matrix, we only consider a limited number of points, called landmarks, and compute the similarity of all data points with the landmarks. Then, we present a definition of the Laplacian matrix of the graph that enable us to perform eigen decomposition efficiently, using a deep autoencoder. The overall complexity of the algorithm for eigen decomposition is $O(np)$, where $n$ is the number of data points and $p$ is the number of landmarks. At last, we evaluate the performance of the algorithm in different experiments.
Tasks
Published	2017-04-07
URL	http://arxiv.org/abs/1704.02345v1
PDF	http://arxiv.org/pdf/1704.02345v1.pdf
PWC	https://paperswithcode.com/paper/fast-spectral-clustering-using-autoencoders
Repo	https://github.com/chenjs12/ML
Framework	none

Dual Rectified Linear Units (DReLUs): A Replacement for Tanh Activation Functions in Quasi-Recurrent Neural Networks


Title	Dual Rectified Linear Units (DReLUs): A Replacement for Tanh Activation Functions in Quasi-Recurrent Neural Networks
Authors	Fréderic Godin, Jonas Degrave, Joni Dambre, Wesley De Neve
Abstract	In this paper, we introduce a novel type of Rectified Linear Unit (ReLU), called a Dual Rectified Linear Unit (DReLU). A DReLU, which comes with an unbounded positive and negative image, can be used as a drop-in replacement for a tanh activation function in the recurrent step of Quasi-Recurrent Neural Networks (QRNNs) (Bradbury et al. (2017)). Similar to ReLUs, DReLUs are less prone to the vanishing gradient problem, they are noise robust, and they induce sparse activations. We independently reproduce the QRNN experiments of Bradbury et al. (2017) and compare our DReLU-based QRNNs with the original tanh-based QRNNs and Long Short-Term Memory networks (LSTMs) on sentiment classification and word-level language modeling. Additionally, we evaluate on character-level language modeling, showing that we are able to stack up to eight QRNN layers with DReLUs, thus making it possible to improve the current state-of-the-art in character-level language modeling over shallow architectures based on LSTMs.
Tasks	Language Modelling, Sentiment Analysis
Published	2017-07-25
URL	http://arxiv.org/abs/1707.08214v2
PDF	http://arxiv.org/pdf/1707.08214v2.pdf
PWC	https://paperswithcode.com/paper/dual-rectified-linear-units-drelus-a
Repo	https://github.com/FredericGodin/QuasiRNN-DReLU
Framework	none

Open Set Logo Detection and Retrieval


Title	Open Set Logo Detection and Retrieval
Authors	Andras Tüzkö, Christian Herrmann, Daniel Manger, Jürgen Beyerer
Abstract	Current logo retrieval research focuses on closed set scenarios. We argue that the logo domain is too large for this strategy and requires an open set approach. To foster research in this direction, a large-scale logo dataset, called Logos in the Wild, is collected and released to the public. A typical open set logo retrieval application is, for example, assessing the effectiveness of advertisement in sports event broadcasts. Given a query sample in shape of a logo image, the task is to find all further occurrences of this logo in a set of images or videos. Currently, common logo retrieval approaches are unsuitable for this task because of their closed world assumption. Thus, an open set logo retrieval method is proposed in this work which allows searching for previously unseen logos by a single query sample. A two stage concept with separate logo detection and comparison is proposed where both modules are based on task specific CNNs. If trained with the Logos in the Wild data, significant performance improvements are observed, especially compared with state-of-the-art closed set approaches.
Tasks
Published	2017-10-30
URL	http://arxiv.org/abs/1710.10891v1
PDF	http://arxiv.org/pdf/1710.10891v1.pdf
PWC	https://paperswithcode.com/paper/open-set-logo-detection-and-retrieval
Repo	https://github.com/ilmonteux/logohunter
Framework	tf