January 27, 2020

3200 words 16 mins read

Paper Group ANR 1338

An Investigation of Data Poisoning Defenses for Online Learning. Few-Shot Learning with Localization in Realistic Settings. Generative Adversarial Imagination for Sample Efficient Deep Reinforcement Learning. Fast Polynomial Approximation of Heat Kernel Convolution on Manifolds and Its Application to Brain Sulcal and Gyral Graph Pattern Analysis. D …

An Investigation of Data Poisoning Defenses for Online Learning


Title	An Investigation of Data Poisoning Defenses for Online Learning
Authors	Yizhen Wang, Somesh Jha, Kamalika Chaudhuri
Abstract	Data poisoning attacks – where an adversary can modify a small fraction of training data, with the goal of forcing the trained classifier to high loss – are an important threat for machine learning in many applications. While a body of prior work has developed attacks and defenses, there is not much general understanding on when various attacks and defenses are effective. In this work, we undertake a rigorous study of defenses against data poisoning for online learning. First, we study four standard defenses in a powerful threat model, and provide conditions under which they can allow or resist rapid poisoning. We then consider a weaker and more realistic threat model, and show that the success of the adversary in the presence of data poisoning defenses there depends on the “ease” of the learning problem.
Tasks	data poisoning
Published	2019-05-28
URL	https://arxiv.org/abs/1905.12121v3
PDF	https://arxiv.org/pdf/1905.12121v3.pdf
PWC	https://paperswithcode.com/paper/an-investigation-of-data-poisoning-defenses
Repo
Framework

Few-Shot Learning with Localization in Realistic Settings


Title	Few-Shot Learning with Localization in Realistic Settings
Authors	Davis Wertheimer, Bharath Hariharan
Abstract	Traditional recognition methods typically require large, artificially-balanced training classes, while few-shot learning methods are tested on artificially small ones. In contrast to both extremes, real world recognition problems exhibit heavy-tailed class distributions, with cluttered scenes and a mix of coarse and fine-grained class distinctions. We show that prior methods designed for few-shot learning do not work out of the box in these challenging conditions, based on a new “meta-iNat” benchmark. We introduce three parameter-free improvements: (a) better training procedures based on adapting cross-validation to meta-learning, (b) novel architectures that localize objects using limited bounding box annotations before classification, and (c) simple parameter-free expansions of the feature space based on bilinear pooling. Together, these improvements double the accuracy of state-of-the-art models on meta-iNat while generalizing to prior benchmarks, complex neural architectures, and settings with substantial domain shift.
Tasks	Few-Shot Learning, Meta-Learning
Published	2019-04-09
URL	https://arxiv.org/abs/1904.08502v2
PDF	https://arxiv.org/pdf/1904.08502v2.pdf
PWC	https://paperswithcode.com/paper/190408502
Repo
Framework

Generative Adversarial Imagination for Sample Efficient Deep Reinforcement Learning


Title	Generative Adversarial Imagination for Sample Efficient Deep Reinforcement Learning
Authors	Kacper Kielak
Abstract	Reinforcement learning has seen great advancements in the past five years. The successful introduction of deep learning in place of more traditional methods allowed reinforcement learning to scale to very complex domains achieving super-human performance in environments like the game of Go or numerous video games. Despite great successes in multiple domains, these new methods suffer from their own issues that make them often inapplicable to the real world problems. Extreme lack of data efficiency, together with huge variance and difficulty in enforcing safety constraints, is one of the three most prominent issues in the field. Usually, millions of data points sampled from the environment are necessary for these algorithms to converge to acceptable policies. This thesis proposes novel Generative Adversarial Imaginative Reinforcement Learning algorithm. It takes advantage of the recent introduction of highly effective generative adversarial models, and Markov property that underpins reinforcement learning setting, to model dynamics of the real environment within the internal imagination module. Rollouts from the imagination are then used to artificially simulate the real environment in a standard reinforcement learning process to avoid, often expensive and dangerous, trial and error in the real environment. Experimental results show that the proposed algorithm more economically utilises experience from the real environment than the current state-of-the-art Rainbow DQN algorithm, and thus makes an important step towards sample efficient deep reinforcement learning.
Tasks	Game of Go
Published	2019-04-30
URL	https://arxiv.org/abs/1904.13255v2
PDF	https://arxiv.org/pdf/1904.13255v2.pdf
PWC	https://paperswithcode.com/paper/generative-adversarial-imagination-for-sample
Repo
Framework

Fast Polynomial Approximation of Heat Kernel Convolution on Manifolds and Its Application to Brain Sulcal and Gyral Graph Pattern Analysis


Title	Fast Polynomial Approximation of Heat Kernel Convolution on Manifolds and Its Application to Brain Sulcal and Gyral Graph Pattern Analysis
Authors	Shih-Gu Huang, Ilwoo Lyu, Anqi Qiu, Moo K. Chung
Abstract	Heat diffusion has been widely used in brain imaging for surface fairing, mesh regularization and cortical data smoothing. Motivated by diffusion wavelets and convolutional neural networks on graphs, we present a new fast and accurate numerical scheme to solve heat diffusion on surface meshes. This is achieved by approximating the heat kernel convolution using high degree orthogonal polynomials in the spectral domain. We also derive the closed-form expression of the spectral decomposition of the Laplace-Beltrami operator and use it to solve heat diffusion on a manifold for the first time. The proposed fast polynomial approximation scheme avoids solving for the eigenfunctions of the Laplace-Beltrami operator, which is computationally costly for large mesh size, and the numerical instability associated with the finite element method based diffusion solvers. The proposed method is applied in localizing the male and female differences in cortical sulcal and gyral graph patterns obtained from MRI in an innovative way. The MATLAB code is available at http://www.stat.wisc.edu/~mchung/chebyshev.
Tasks
Published	2019-11-07
URL	https://arxiv.org/abs/1911.02721v2
PDF	https://arxiv.org/pdf/1911.02721v2.pdf
PWC	https://paperswithcode.com/paper/fast-polynomial-approximation-of-heat
Repo
Framework

Document Sub-structure in Neural Machine Translation


Title	Document Sub-structure in Neural Machine Translation
Authors	Radina Dobreva, Jie Zhou, Rachel Bawden
Abstract	Current approaches to machine translation (MT) either translate sentences in isolation, disregarding the context they appear in, or model context at the level of the full document, without a notion of any internal structure the document may have. In this work we consider the fact that documents are rarely homogeneous blocks of text, but rather consist of parts covering different topics. Some documents, such as biographies and encyclopedia entries, have highly predictable, regular structures in which sections are characterised by different topics. We draw inspiration from Louis and Webber (2014) who use this information to improve statistical MT and transfer their proposal into the framework of neural MT. We compare two different methods of including information about the topic of the section within which each sentence is found: one using side constraints and the other using a cache-based model. We create and release the data on which we run our experiments - parallel corpora for three language pairs (Chinese-English, French-English, Bulgarian-English) from Wikipedia biographies, which we extract automatically, preserving the boundaries of sections within the articles.
Tasks	Machine Translation
Published	2019-12-13
URL	https://arxiv.org/abs/1912.06598v2
PDF	https://arxiv.org/pdf/1912.06598v2.pdf
PWC	https://paperswithcode.com/paper/document-sub-structure-in-neural-machine
Repo
Framework

Empirical Differential Privacy


Title	Empirical Differential Privacy
Authors	Paul Burchard, Anthony Daoud
Abstract	We show how to achieve differential privacy with no or reduced added noise, based on the empirical noise in the data itself. Unlike previous works on noiseless privacy, the empirical viewpoint avoids making any explicit assumptions about the random process generating the data.
Tasks
Published	2019-10-28
URL	https://arxiv.org/abs/1910.12820v3
PDF	https://arxiv.org/pdf/1910.12820v3.pdf
PWC	https://paperswithcode.com/paper/empirical-differential-privacy
Repo
Framework

PFML-based Semantic BCI Agent for Game of Go Learning and Prediction


Title	PFML-based Semantic BCI Agent for Game of Go Learning and Prediction
Authors	Chang-Shing Lee, Mei-Hui Wang, Li-Wei Ko, Bo-Yu Tsai, Yi-Lin Tsai, Sheng-Chi Yang, Lu-An Lin, Yi-Hsiu Lee, Hirofumi Ohashi, Naoyuki Kubota, Nan Shuo
Abstract	This paper presents a semantic brain computer interface (BCI) agent with particle swarm optimization (PSO) based on a Fuzzy Markup Language (FML) for Go learning and prediction applications. Additionally, we also establish an Open Go Darkforest (OGD) cloud platform with Facebook AI research (FAIR) open source Darkforest and ELF OpenGo AI bots. The Japanese robot Palro will simultaneously predict the move advantage in the board game Go to the Go players for reference or learning. The proposed semantic BCI agent operates efficiently by the human-based BCI data from their brain waves and machine-based game data from the prediction of the OGD cloud platform for optimizing the parameters between humans and machines. Experimental results show that the proposed human and smart machine co-learning mechanism performs favorably. We hope to provide students with a better online learning environment, combining different kinds of handheld devices, robots, or computer equipment, to achieve a desired and intellectual learning goal in the future.
Tasks	Game of Go
Published	2019-01-10
URL	http://arxiv.org/abs/1901.02999v1
PDF	http://arxiv.org/pdf/1901.02999v1.pdf
PWC	https://paperswithcode.com/paper/pfml-based-semantic-bci-agent-for-game-of-go
Repo
Framework

Learning-based Real-time Detection of Intrinsic Reflectional Symmetry


Title	Learning-based Real-time Detection of Intrinsic Reflectional Symmetry
Authors	Yi-Ling Qiao, Lin Gao, Shu-Zhi Liu, Ligang Liu, Yu-Kun Lai, Xilin Chen
Abstract	Reflectional symmetry is ubiquitous in nature. While extrinsic reflectional symmetry can be easily parametrized and detected, intrinsic symmetry is much harder due to the high solution space. Previous works usually solve this problem by voting or sampling, which suffer from high computational cost and randomness. In this paper, we propose \YL{a} learning-based approach to intrinsic reflectional symmetry detection. Instead of directly finding symmetric point pairs, we parametrize this self-isometry using a functional map matrix, which can be easily computed given the signs of Laplacian eigenfunctions under the symmetric mapping. Therefore, we train a novel deep neural network to predict the sign of each eigenfunction under symmetry, which in addition takes the first few eigenfunctions as intrinsic features to characterize the mesh while avoiding coping with the connectivity explicitly. Our network aims at learning the global property of functions, and consequently converts the problem defined on the manifold to the functional domain. By disentangling the prediction of the matrix into separated basis, our method generalizes well to new shapes and is invariant under perturbation of eigenfunctions. Through extensive experiments, we demonstrate the robustness of our method in challenging cases, including different topology and incomplete shapes with holes. By avoiding random sampling, our learning-based algorithm is over 100 times faster than state-of-the-art methods, and meanwhile, is more robust, achieving higher correspondence accuracy in commonly used metrics.
Tasks
Published	2019-11-01
URL	https://arxiv.org/abs/1911.00189v1
PDF	https://arxiv.org/pdf/1911.00189v1.pdf
PWC	https://paperswithcode.com/paper/learning-based-real-time-detection-of
Repo
Framework

Fast and Accurate Knowledge-Aware Document Representation Enhancement for News Recommendations


Title	Fast and Accurate Knowledge-Aware Document Representation Enhancement for News Recommendations
Authors	Danyang Liu, Jianxun Lian, Ying Qiao, Jiun-Hung Chen, Guangzhong Sun, Xing Xie
Abstract	Knowledge graph contains well-structured external information and has shown to be useful for recommender systems. Most existing knowledge-aware methods assume that the item from recommender systems can be linked to an entity in a knowledge graph, thus item embeddings can be better learned by jointly modeling of both recommender systems and a knowledge graph. However, this is not the situation for news recommendation, where items, namely news articles, are in fact related to a collection of knowledge entities. The importance score and semantic information of entities in one article differ from each other, which depend on the topic of the article and relations among co-occurred entities. How to fully utilize these entities for better news recommendation service is non-trivial. In this paper, we propose a fast and effective knowledge-aware representation enhancement model for improving news document understanding. The model, named \emph{KRED}, consists of three layers: (1) an entity representation layer; (2) a context embedding layer; and (3) an information distillation layer. An entity is represented by the embeddings of itself and its surrounding entities. The context embedding layer is designed to distinguish dynamic context of different entities such as frequency, category and position. The information distillation layer will aggregate the entity embeddings under the guidance of the original document vector, transforming the document vector into a new one. We have conduct extensive experiments on a real-world news reading dataset. The results demonstrate that our proposed model greatly benefits a variety of news recommendation tasks, including personalized news recommendation, article category classification, article popularity prediction and local news detection.
Tasks	Entity Embeddings, Recommendation Systems
Published	2019-10-25
URL	https://arxiv.org/abs/1910.11494v1
PDF	https://arxiv.org/pdf/1910.11494v1.pdf
PWC	https://paperswithcode.com/paper/fast-and-accurate-knowledge-aware-document
Repo
Framework

BAG: Bi-directional Attention Entity Graph Convolutional Network for Multi-hop Reasoning Question Answering


Title	BAG: Bi-directional Attention Entity Graph Convolutional Network for Multi-hop Reasoning Question Answering
Authors	Yu Cao, Meng Fang, Dacheng Tao
Abstract	Multi-hop reasoning question answering requires deep comprehension of relationships between various documents and queries. We propose a Bi-directional Attention Entity Graph Convolutional Network (BAG), leveraging relationships between nodes in an entity graph and attention information between a query and the entity graph, to solve this task. Graph convolutional networks are used to obtain a relation-aware representation of nodes for entity graphs built from documents with multi-level features. Bidirectional attention is then applied on graphs and queries to generate a query-aware nodes representation, which will be used for the final prediction. Experimental evaluation shows BAG achieves state-of-the-art accuracy performance on the QAngaroo WIKIHOP dataset.
Tasks	Question Answering
Published	2019-04-10
URL	http://arxiv.org/abs/1904.04969v1
PDF	http://arxiv.org/pdf/1904.04969v1.pdf
PWC	https://paperswithcode.com/paper/bag-bi-directional-attention-entity-graph
Repo
Framework

Aligning Cross-Lingual Entities with Multi-Aspect Information


Title	Aligning Cross-Lingual Entities with Multi-Aspect Information
Authors	Hsiu-Wei Yang, Yanyan Zou, Peng Shi, Wei Lu, Jimmy Lin, Xu Sun
Abstract	Multilingual knowledge graphs (KGs), such as YAGO and DBpedia, represent entities in different languages. The task of cross-lingual entity alignment is to match entities in a source language with their counterparts in target languages. In this work, we investigate embedding-based approaches to encode entities from multilingual KGs into the same vector space, where equivalent entities are close to each other. Specifically, we apply graph convolutional networks (GCNs) to combine multi-aspect information of entities, including topological connections, relations, and attributes of entities, to learn entity embeddings. To exploit the literal descriptions of entities expressed in different languages, we propose two uses of a pretrained multilingual BERT model to bridge cross-lingual gaps. We further propose two strategies to integrate GCN-based and BERT-based modules to boost performance. Extensive experiments on two benchmark datasets demonstrate that our method significantly outperforms existing systems.
Tasks	Entity Alignment, Entity Embeddings, Knowledge Graphs
Published	2019-10-15
URL	https://arxiv.org/abs/1910.06575v1
PDF	https://arxiv.org/pdf/1910.06575v1.pdf
PWC	https://paperswithcode.com/paper/aligning-cross-lingual-entities-with-multi
Repo
Framework

Substra: a framework for privacy-preserving, traceable and collaborative Machine Learning


Title	Substra: a framework for privacy-preserving, traceable and collaborative Machine Learning
Authors	Mathieu N Galtier, Camille Marini
Abstract	Machine learning is promising, but it often needs to process vast amounts of sensitive data which raises concerns about privacy. In this white-paper, we introduce Substra, a distributed framework for privacy-preserving, traceable and collaborative Machine Learning. Substra gathers data providers and algorithm designers into a network of nodes that can train models on demand but under advanced permission regimes. To guarantee data privacy, Substra implements distributed learning: the data never leave their nodes; only algorithms, predictive models and non-sensitive metadata are exchanged on the network. The computations are orchestrated by a Distributed Ledger Technology which guarantees traceability and authenticity of information without needing to trust a third party. Although originally developed for Healthcare applications, Substra is not data, algorithm or programming language specific. It supports many types of computation plans including parallel computation plan commonly used in Federated Learning. With appropriate guidelines, it can be deployed for numerous Machine Learning use-cases with data or algorithm providers where trust is limited.
Tasks
Published	2019-10-25
URL	https://arxiv.org/abs/1910.11567v1
PDF	https://arxiv.org/pdf/1910.11567v1.pdf
PWC	https://paperswithcode.com/paper/substra-a-framework-for-privacy-preserving
Repo
Framework

Detection of Community Structures in Networks with Nodal Features based on Generative Probabilistic Approach


Title	Detection of Community Structures in Networks with Nodal Features based on Generative Probabilistic Approach
Authors	Hadi Zare, Mahdi Hajiabadi, Mahdi Jalili
Abstract	Community detection is considered as a fundamental task in analyzing social networks. Even though many techniques have been proposed for community detection, most of them are based exclusively on the connectivity structures. However, there are node features in real networks, such as gender types in social networks, feeding behavior in ecological networks, and location on e-trading networks, that can be further leveraged with the network structure to attain more accurate community detection methods. We propose a novel probabilistic graphical model to detect communities by taking into account both network structure and nodes’ features. The proposed approach learns the relevant features of communities through a generative probabilistic model without any prior assumption on the communities. Furthermore, the model is capable of determining the strength of node features and structural elements of the networks on shaping the communities. The effectiveness of the proposed approach over the state-of-the-art algorithms is revealed on synthetic and benchmark networks.
Tasks	Community Detection
Published	2019-12-24
URL	https://arxiv.org/abs/1912.11420v1
PDF	https://arxiv.org/pdf/1912.11420v1.pdf
PWC	https://paperswithcode.com/paper/detection-of-community-structures-in-networks
Repo
Framework

REPLAB: A Reproducible Low-Cost Arm Benchmark Platform for Robotic Learning


Title	REPLAB: A Reproducible Low-Cost Arm Benchmark Platform for Robotic Learning
Authors	Brian Yang, Jesse Zhang, Vitchyr Pong, Sergey Levine, Dinesh Jayaraman
Abstract	Standardized evaluation measures have aided in the progress of machine learning approaches in disciplines such as computer vision and machine translation. In this paper, we make the case that robotic learning would also benefit from benchmarking, and present the “REPLAB” platform for benchmarking vision-based manipulation tasks. REPLAB is a reproducible and self-contained hardware stack (robot arm, camera, and workspace) that costs about 2000 USD, occupies a cuboid of size 70x40x60 cm, and permits full assembly within a few hours. Through this low-cost, compact design, REPLAB aims to drive wide participation by lowering the barrier to entry into robotics and to enable easy scaling to many robots. We envision REPLAB as a framework for reproducible research across manipulation tasks, and as a step in this direction, we define a template for a grasping benchmark consisting of a task definition, evaluation protocol, performance measures, and a dataset of 92k grasp attempts. We implement, evaluate, and analyze several previously proposed grasping approaches to establish baselines for this benchmark. Finally, we also implement and evaluate a deep reinforcement learning approach for 3D reaching tasks on our REPLAB platform. Project page with assembly instructions, code, and videos: https://goo.gl/5F9dP4.
Tasks	Machine Translation
Published	2019-05-17
URL	https://arxiv.org/abs/1905.07447v1
PDF	https://arxiv.org/pdf/1905.07447v1.pdf
PWC	https://paperswithcode.com/paper/replab-a-reproducible-low-cost-arm-benchmark
Repo
Framework

Referring Expression Grounding by Marginalizing Scene Graph Likelihood


Title	Referring Expression Grounding by Marginalizing Scene Graph Likelihood
Authors	Daqing Liu, Hanwang Zhang, Zheng-Jun Zha, Fanglin Wang
Abstract	We focus on the task of grounding referring expressions in images, e.g., localizing “the white truck in front of a yellow one”. To resolve this task fundamentally, one should first find out the contextual objects (e.g., the “yellow” truck) and then exploit them to disambiguate the referent from other similar objects, by using the attributes and relationships (e.g., “white”, “yellow”, “in front of”). However, it is extremely challenging to train such a model as the ground-truth of the contextual objects and their relationships are usually missing due to the prohibitive annotation cost. Therefore, nearly all existing methods attempt to evade the above joint grounding and reasoning process, but resort to a holistic association between the sentence and region feature. As a result, they suffer from heavy parameters of fully-connected layers, poor interpretability, and limited generalization to unseen expressions. In this paper, we tackle this challenge by training and inference with the proposed Marginalized Scene Graph Likelihood (MSGL). Specifically, we use scene graph: a graphical representation parsed from the referring expression, where the nodes are objects with attributes and the edges are relationships. Thanks to the conditional random field (CRF) built on scene graph, we can ground every object to its corresponding region, and perform reasoning with the unlabeled contexts by marginalizing out them using the sum-product belief propagation. Overall, our proposed MSGL is effective and interpretable, e.g., on three benchmarks, MSGL consistently outperforms the state-of-the-arts while offers a complete grounding of all the objects in a sentence.
Tasks
Published	2019-06-09
URL	https://arxiv.org/abs/1906.03561v1
PDF	https://arxiv.org/pdf/1906.03561v1.pdf
PWC	https://paperswithcode.com/paper/referring-expression-grounding-by
Repo
Framework