May 6, 2019

3063 words 15 mins read

Paper Group ANR 183

Learning Dexterous Manipulation Policies from Experience and Imitation. Oracle Inequalities for High-dimensional Prediction. Unsupervised Discovery of El Nino Using Causal Feature Learning on Microlevel Climate Data. Enabling Cognitive Intelligence Queries in Relational Databases using Low-dimensional Word Embeddings. On-Average KL-Privacy and its …

Learning Dexterous Manipulation Policies from Experience and Imitation


Title	Learning Dexterous Manipulation Policies from Experience and Imitation
Authors	Vikash Kumar, Abhishek Gupta, Emanuel Todorov, Sergey Levine
Abstract	We explore learning-based approaches for feedback control of a dexterous five-finger hand performing non-prehensile manipulation. First, we learn local controllers that are able to perform the task starting at a predefined initial state. These controllers are constructed using trajectory optimization with respect to locally-linear time-varying models learned directly from sensor data. In some cases, we initialize the optimizer with human demonstrations collected via teleoperation in a virtual environment. We demonstrate that such controllers can perform the task robustly, both in simulation and on the physical platform, for a limited range of initial conditions around the trained starting state. We then consider two interpolation methods for generalizing to a wider range of initial conditions: deep learning, and nearest neighbors. We find that nearest neighbors achieve higher performance. Nevertheless, the neural network has its advantages: it uses only tactile and proprioceptive feedback but no visual feedback about the object (i.e. it performs the task blind) and learns a time-invariant policy. In contrast, the nearest neighbors method switches between time-varying local controllers based on the proximity of initial object states sensed via motion capture. While both generalization methods leave room for improvement, our work shows that (i) local trajectory-based controllers for complex non-prehensile manipulation tasks can be constructed from surprisingly small amounts of training data, and (ii) collections of such controllers can be interpolated to form more global controllers. Results are summarized in the supplementary video: https://youtu.be/E0wmO6deqjo
Tasks	Motion Capture
Published	2016-11-15
URL	http://arxiv.org/abs/1611.05095v1
PDF	http://arxiv.org/pdf/1611.05095v1.pdf
PWC	https://paperswithcode.com/paper/learning-dexterous-manipulation-policies-from
Repo
Framework

Oracle Inequalities for High-dimensional Prediction


Title	Oracle Inequalities for High-dimensional Prediction
Authors	Johannes Lederer, Lu Yu, Irina Gaynanova
Abstract	The abundance of high-dimensional data in the modern sciences has generated tremendous interest in penalized estimators such as the lasso, scaled lasso, square-root lasso, elastic net, and many others. In this paper, we establish a general oracle inequality for prediction in high-dimensional linear regression with such methods. Since the proof relies only on convexity and continuity arguments, the result holds irrespective of the design matrix and applies to a wide range of penalized estimators. Overall, the bound demonstrates that generic estimators can provide consistent prediction with any design matrix. From a practical point of view, the bound can help to identify the potential of specific estimators, and they can help to get a sense of the prediction accuracy in a given application.
Tasks
Published	2016-08-01
URL	http://arxiv.org/abs/1608.00624v2
PDF	http://arxiv.org/pdf/1608.00624v2.pdf
PWC	https://paperswithcode.com/paper/oracle-inequalities-for-high-dimensional-1
Repo
Framework

Unsupervised Discovery of El Nino Using Causal Feature Learning on Microlevel Climate Data


Title	Unsupervised Discovery of El Nino Using Causal Feature Learning on Microlevel Climate Data
Authors	Krzysztof Chalupka, Tobias Bischoff, Pietro Perona, Frederick Eberhardt
Abstract	We show that the climate phenomena of El Nino and La Nina arise naturally as states of macro-variables when our recent causal feature learning framework (Chalupka 2015, Chalupka 2016) is applied to micro-level measures of zonal wind (ZW) and sea surface temperatures (SST) taken over the equatorial band of the Pacific Ocean. The method identifies these unusual climate states on the basis of the relation between ZW and SST patterns without any input about past occurrences of El Nino or La Nina. The simpler alternatives of (i) clustering the SST fields while disregarding their relationship with ZW patterns, or (ii) clustering the joint ZW-SST patterns, do not discover El Nino. We discuss the degree to which our method supports a causal interpretation and use a low-dimensional toy example to explain its success over other clustering approaches. Finally, we propose a new robust and scalable alternative to our original algorithm (Chalupka 2016), which circumvents the need for high-dimensional density learning.
Tasks
Published	2016-05-30
URL	http://arxiv.org/abs/1605.09370v1
PDF	http://arxiv.org/pdf/1605.09370v1.pdf
PWC	https://paperswithcode.com/paper/unsupervised-discovery-of-el-nino-using
Repo
Framework

Enabling Cognitive Intelligence Queries in Relational Databases using Low-dimensional Word Embeddings


Title	Enabling Cognitive Intelligence Queries in Relational Databases using Low-dimensional Word Embeddings
Authors	Rajesh Bordawekar, Oded Shmueli
Abstract	We apply distributed language embedding methods from Natural Language Processing to assign a vector to each database entity associated token (for example, a token may be a word occurring in a table row, or the name of a column). These vectors, of typical dimension 200, capture the meaning of tokens based on the contexts in which the tokens appear together. To form vectors, we apply a learning method to a token sequence derived from the database. We describe various techniques for extracting token sequences from a database. The techniques differ in complexity, in the token sequences they output and in the database information used (e.g., foreign keys). The vectors can be used to algebraically quantify semantic relationships between the tokens such as similarities and analogies. Vectors enable a dual view of the data: relational and (meaningful rather than purely syntactical) text. We introduce and explore a new class of queries called cognitive intelligence (CI) queries that extract information from the database based, in part, on the relationships encoded by vectors. We have implemented a prototype system on top of Spark to exhibit the power of CI queries. Here, CI queries are realized via SQL UDFs. This power goes far beyond text extensions to relational systems due to the information encoded in vectors. We also consider various extensions to the basic scheme, including using a collection of views derived from the database to focus on a domain of interest, utilizing vectors and/or text from external sources, maintaining vectors as the database evolves and exploring a database without utilizing its schema. For the latter, we consider minimal extensions to SQL to vastly improve query expressiveness.
Tasks	Word Embeddings
Published	2016-03-23
URL	http://arxiv.org/abs/1603.07185v1
PDF	http://arxiv.org/pdf/1603.07185v1.pdf
PWC	https://paperswithcode.com/paper/enabling-cognitive-intelligence-queries-in
Repo
Framework

On-Average KL-Privacy and its equivalence to Generalization for Max-Entropy Mechanisms


Title	On-Average KL-Privacy and its equivalence to Generalization for Max-Entropy Mechanisms
Authors	Yu-Xiang Wang, Jing Lei, Stephen E. Fienberg
Abstract	We define On-Average KL-Privacy and present its properties and connections to differential privacy, generalization and information-theoretic quantities including max-information and mutual information. The new definition significantly weakens differential privacy, while preserving its minimalistic design features such as composition over small group and multiple queries as well as closeness to post-processing. Moreover, we show that On-Average KL-Privacy is equivalent to generalization for a large class of commonly-used tools in statistics and machine learning that samples from Gibbs distributions—a class of distributions that arises naturally from the maximum entropy principle. In addition, a byproduct of our analysis yields a lower bound for generalization error in terms of mutual information which reveals an interesting interplay with known upper bounds that use the same quantity.
Tasks
Published	2016-05-08
URL	http://arxiv.org/abs/1605.02277v1
PDF	http://arxiv.org/pdf/1605.02277v1.pdf
PWC	https://paperswithcode.com/paper/on-average-kl-privacy-and-its-equivalence-to
Repo
Framework

Learning molecular energies using localized graph kernels


Title	Learning molecular energies using localized graph kernels
Authors	G. Ferré, T. Haut, K. Barros
Abstract	Recent machine learning methods make it possible to model potential energy of atomic configurations with chemical-level accuracy (as calculated from ab-initio calculations) and at speeds suitable for molecular dynam- ics simulation. Best performance is achieved when the known physical constraints are encoded in the machine learning models. For example, the atomic energy is invariant under global translations and rotations, it is also invariant to permutations of same-species atoms. Although simple to state, these symmetries are complicated to encode into machine learning algorithms. In this paper, we present a machine learning approach based on graph theory that naturally incorporates translation, rotation, and permutation symmetries. Specifically, we use a random walk graph kernel to measure the similarity of two adjacency matrices, each of which represents a local atomic environment. This Graph Approximated Energy (GRAPE) approach is flexible and admits many possible extensions. We benchmark a simple version of GRAPE by predicting atomization energies on a standard dataset of organic molecules.
Tasks
Published	2016-12-01
URL	http://arxiv.org/abs/1612.00193v2
PDF	http://arxiv.org/pdf/1612.00193v2.pdf
PWC	https://paperswithcode.com/paper/learning-molecular-energies-using-localized
Repo
Framework

A data augmentation methodology for training machine/deep learning gait recognition algorithms


Title	A data augmentation methodology for training machine/deep learning gait recognition algorithms
Authors	Christoforos C. Charalambous, Anil A. Bharath
Abstract	There are several confounding factors that can reduce the accuracy of gait recognition systems. These factors can reduce the distinctiveness, or alter the features used to characterise gait, they include variations in clothing, lighting, pose and environment, such as the walking surface. Full invariance to all confounding factors is challenging in the absence of high-quality labelled training data. We introduce a simulation-based methodology and a subject-specific dataset which can be used for generating synthetic video frames and sequences for data augmentation. With this methodology, we generated a multi-modal dataset. In addition, we supply simulation files that provide the ability to simultaneously sample from several confounding variables. The basis of the data is real motion capture data of subjects walking and running on a treadmill at different speeds. Results from gait recognition experiments suggest that information about the identity of subjects is retained within synthetically generated examples. The dataset and methodology allow studies into fully-invariant identity recognition spanning a far greater number of observation conditions than would otherwise be possible.
Tasks	Data Augmentation, Gait Recognition, Motion Capture
Published	2016-10-24
URL	http://arxiv.org/abs/1610.07570v1
PDF	http://arxiv.org/pdf/1610.07570v1.pdf
PWC	https://paperswithcode.com/paper/a-data-augmentation-methodology-for-training
Repo
Framework

Learning Shared Representations in Multi-task Reinforcement Learning


Title	Learning Shared Representations in Multi-task Reinforcement Learning
Authors	Diana Borsa, Thore Graepel, John Shawe-Taylor
Abstract	We investigate a paradigm in multi-task reinforcement learning (MT-RL) in which an agent is placed in an environment and needs to learn to perform a series of tasks, within this space. Since the environment does not change, there is potentially a lot of common ground amongst tasks and learning to solve them individually seems extremely wasteful. In this paper, we explicitly model and learn this shared structure as it arises in the state-action value space. We will show how one can jointly learn optimal value-functions by modifying the popular Value-Iteration and Policy-Iteration procedures to accommodate this shared representation assumption and leverage the power of multi-task supervised learning. Finally, we demonstrate that the proposed model and training procedures, are able to infer good value functions, even under low samples regimes. In addition to data efficiency, we will show in our analysis, that learning abstractions of the state space jointly across tasks leads to more robust, transferable representations with the potential for better generalization. this shared representation assumption and leverage the power of multi-task supervised learning. Finally, we demonstrate that the proposed model and training procedures, are able to infer good value functions, even under low samples regimes. In addition to data efficiency, we will show in our analysis, that learning abstractions of the state space jointly across tasks leads to more robust, transferable representations with the potential for better generalization.
Tasks
Published	2016-03-07
URL	http://arxiv.org/abs/1603.02041v1
PDF	http://arxiv.org/pdf/1603.02041v1.pdf
PWC	https://paperswithcode.com/paper/learning-shared-representations-in-multi-task
Repo
Framework

The Microsoft 2016 Conversational Speech Recognition System


Title	The Microsoft 2016 Conversational Speech Recognition System
Authors	W. Xiong, J. Droppo, X. Huang, F. Seide, M. Seltzer, A. Stolcke, D. Yu, G. Zweig
Abstract	We describe Microsoft’s conversational speech recognition system, in which we combine recent developments in neural-network-based acoustic and language modeling to advance the state of the art on the Switchboard recognition task. Inspired by machine learning ensemble techniques, the system uses a range of convolutional and recurrent neural networks. I-vector modeling and lattice-free MMI training provide significant gains for all acoustic model architectures. Language model rescoring with multiple forward and backward running RNNLMs, and word posterior-based system combination provide a 20% boost. The best single system uses a ResNet architecture acoustic model with RNNLM rescoring, and achieves a word error rate of 6.9% on the NIST 2000 Switchboard task. The combined system has an error rate of 6.2%, representing an improvement over previously reported results on this benchmark task.
Tasks	Language Modelling, Speech Recognition
Published	2016-09-12
URL	http://arxiv.org/abs/1609.03528v2
PDF	http://arxiv.org/pdf/1609.03528v2.pdf
PWC	https://paperswithcode.com/paper/the-microsoft-2016-conversational-speech
Repo
Framework

A Framework for Estimating Long Term Driver Behavior


Title	A Framework for Estimating Long Term Driver Behavior
Authors	Vijay Gadepally, Ashok Krishnamurthy
Abstract	The authors present a cyber-physical systems study on the estimation of driver behavior in autonomous vehicles and vehicle safety systems. Extending upon previous work, the approach described is suitable for the long term estimation and tracking of autonomous vehicle behavior. The proposed system makes use of a previously defined Hybrid State System and Hidden Markov Model (HSS+HMM) system which has provided good results for driver behavior estimation. The HSS+HMM system utilizes the hybrid characteristics of decision-behavior coupling of many systems such as the driver and the vehicle, uses Kalman Filter estimates of observable parameters to track the instantaneous continuous state, and estimates the most likely driver state. The HSS+HMM system is encompassed in a HSS structure and inter-system connectivity is determined by using Signal Processing and Pattern Recognition techniques. The proposed method is suitable for scenarios that involve unknown decisions of other individuals, such as lane changes or intersection precedence/access. The long term driver behavior estimation system involves an extended HSS+HMM structure that is capable of including external information in the estimation process. Through the grafting and pruning of metastates, the HSS+HMM system can be dynamically updated to best represent driver choices given external information. Three application examples are also provided to elucidate the theoretical system.
Tasks	Autonomous Vehicles
Published	2016-07-11
URL	http://arxiv.org/abs/1607.03189v1
PDF	http://arxiv.org/pdf/1607.03189v1.pdf
PWC	https://paperswithcode.com/paper/a-framework-for-estimating-long-term-driver
Repo
Framework

Lost and Found: Detecting Small Road Hazards for Self-Driving Vehicles


Title	Lost and Found: Detecting Small Road Hazards for Self-Driving Vehicles
Authors	Peter Pinggera, Sebastian Ramos, Stefan Gehrig, Uwe Franke, Carsten Rother, Rudolf Mester
Abstract	Detecting small obstacles on the road ahead is a critical part of the driving task which has to be mastered by fully autonomous cars. In this paper, we present a method based on stereo vision to reliably detect such obstacles from a moving vehicle. The proposed algorithm performs statistical hypothesis tests in disparity space directly on stereo image data, assessing freespace and obstacle hypotheses on independent local patches. This detection approach does not depend on a global road model and handles both static and moving obstacles. For evaluation, we employ a novel lost-cargo image sequence dataset comprising more than two thousand frames with pixelwise annotations of obstacle and free-space and provide a thorough comparison to several stereo-based baseline methods. The dataset will be made available to the community to foster further research on this important topic. The proposed approach outperforms all considered baselines in our evaluations on both pixel and object level and runs at frame rates of up to 20 Hz on 2 mega-pixel stereo imagery. Small obstacles down to the height of 5 cm can successfully be detected at 20 m distance at low false positive rates.
Tasks
Published	2016-09-15
URL	http://arxiv.org/abs/1609.04653v1
PDF	http://arxiv.org/pdf/1609.04653v1.pdf
PWC	https://paperswithcode.com/paper/lost-and-found-detecting-small-road-hazards
Repo
Framework


Title	Towards the Modeling of Behavioral Trajectories of Users in Online Social Media
Authors	Alessandro Bessi
Abstract	In this paper, we introduce a methodology that allows to model behavioral trajectories of users in online social media. First, we illustrate how to leverage the probabilistic framework provided by Hidden Markov Models (HMMs) to represent users by embedding the temporal sequences of actions they performed online. We then derive a model-based distance between trained HMMs, and we use spectral clustering to find homogeneous clusters of users showing similar behavioral trajectories. To provide platform-agnostic results, we apply the proposed approach to two different online social media — i.e. Facebook and YouTube. We conclude discussing merits and limitations of our approach as well as future and promising research directions.
Tasks
Published	2016-11-17
URL	http://arxiv.org/abs/1611.05778v2
PDF	http://arxiv.org/pdf/1611.05778v2.pdf
PWC	https://paperswithcode.com/paper/towards-the-modeling-of-behavioral
Repo
Framework

Conspiracies between Learning Algorithms, Circuit Lower Bounds and Pseudorandomness


Title	Conspiracies between Learning Algorithms, Circuit Lower Bounds and Pseudorandomness
Authors	Igor C. Oliveira, Rahul Santhanam
Abstract	We prove several results giving new and stronger connections between learning, circuit lower bounds and pseudorandomness. Among other results, we show a generic learning speedup lemma, equivalences between various learning models in the exponential time and subexponential time regimes, a dichotomy between learning and pseudorandomness, consequences of non-trivial learning for circuit lower bounds, Karp-Lipton theorems for probabilistic exponential time, and NC$^1$-hardness for the Minimum Circuit Size Problem.
Tasks
Published	2016-11-03
URL	http://arxiv.org/abs/1611.01190v1
PDF	http://arxiv.org/pdf/1611.01190v1.pdf
PWC	https://paperswithcode.com/paper/conspiracies-between-learning-algorithms
Repo
Framework

Max-Norm Optimization for Robust Matrix Recovery


Title	Max-Norm Optimization for Robust Matrix Recovery
Authors	Ethan X. Fang, Han Liu, Kim-Chuan Toh, Wen-Xin Zhou
Abstract	This paper studies the matrix completion problem under arbitrary sampling schemes. We propose a new estimator incorporating both max-norm and nuclear-norm regularization, based on which we can conduct efficient low-rank matrix recovery using a random subset of entries observed with additive noise under general non-uniform and unknown sampling distributions. This method significantly relaxes the uniform sampling assumption imposed for the widely used nuclear-norm penalized approach, and makes low-rank matrix recovery feasible in more practical settings. Theoretically, we prove that the proposed estimator achieves fast rates of convergence under different settings. Computationally, we propose an alternating direction method of multipliers algorithm to efficiently compute the estimator, which bridges a gap between theory and practice of machine learning methods with max-norm regularization. Further, we provide thorough numerical studies to evaluate the proposed method using both simulated and real datasets.
Tasks	Matrix Completion
Published	2016-09-24
URL	http://arxiv.org/abs/1609.07664v1
PDF	http://arxiv.org/pdf/1609.07664v1.pdf
PWC	https://paperswithcode.com/paper/max-norm-optimization-for-robust-matrix
Repo
Framework

Bi-directional LSTM Recurrent Neural Network for Chinese Word Segmentation


Title	Bi-directional LSTM Recurrent Neural Network for Chinese Word Segmentation
Authors	Yushi Yao, Zheng Huang
Abstract	Recurrent neural network(RNN) has been broadly applied to natural language processing(NLP) problems. This kind of neural network is designed for modeling sequential data and has been testified to be quite efficient in sequential tagging tasks. In this paper, we propose to use bi-directional RNN with long short-term memory(LSTM) units for Chinese word segmentation, which is a crucial preprocess task for modeling Chinese sentences and articles. Classical methods focus on designing and combining hand-craft features from context, whereas bi-directional LSTM network(BLSTM) does not need any prior knowledge or pre-designing, and it is expert in keeping the contextual information in both directions. Experiment result shows that our approach gets state-of-the-art performance in word segmentation on both traditional Chinese datasets and simplified Chinese datasets.
Tasks	Chinese Word Segmentation
Published	2016-02-16
URL	http://arxiv.org/abs/1602.04874v1
PDF	http://arxiv.org/pdf/1602.04874v1.pdf
PWC	https://paperswithcode.com/paper/bi-directional-lstm-recurrent-neural-network
Repo
Framework