May 6, 2019

3244 words 16 mins read

Paper Group ANR 428

Robust Automated Human Activity Recognition and its Application to Sleep Research. Loss is its own Reward: Self-Supervision for Reinforcement Learning. Corporate Disruption in the Science of Machine Learning. Modular Deep Q Networks for Sim-to-real Transfer of Visuo-motor Policies. Review of Fall Detection Techniques: A Data Availability Perspectiv …

Robust Automated Human Activity Recognition and its Application to Sleep Research


Title	Robust Automated Human Activity Recognition and its Application to Sleep Research
Authors	Aarti Sathyanarayana, Ferda Ofli, Luis Fernandes-Luque, Jaideep Srivastava, Ahmed Elmagarmid, Teresa Arora, Shahrad Taheri
Abstract	Human Activity Recognition (HAR) is a powerful tool for understanding human behaviour. Applying HAR to wearable sensors can provide new insights by enriching the feature set in health studies, and enhance the personalisation and effectiveness of health, wellness, and fitness applications. Wearable devices provide an unobtrusive platform for user monitoring, and due to their increasing market penetration, feel intrinsic to the wearer. The integration of these devices in daily life provide a unique opportunity for understanding human health and wellbeing. This is referred to as the “quantified self” movement. The analyses of complex health behaviours such as sleep, traditionally require a time-consuming manual interpretation by experts. This manual work is necessary due to the erratic periodicity and persistent noisiness of human behaviour. In this paper, we present a robust automated human activity recognition algorithm, which we call RAHAR. We test our algorithm in the application area of sleep research by providing a novel framework for evaluating sleep quality and examining the correlation between the aforementioned and an individual’s physical activity. Our results improve the state-of-the-art procedure in sleep research by 15 percent for area under ROC and by 30 percent for F1 score on average. However, application of RAHAR is not limited to sleep analysis and can be used for understanding other health problems such as obesity, diabetes, and cardiac diseases.
Tasks	Activity Recognition, Human Activity Recognition, Sleep Quality
Published	2016-07-17
URL	http://arxiv.org/abs/1607.04867v2
PDF	http://arxiv.org/pdf/1607.04867v2.pdf
PWC	https://paperswithcode.com/paper/robust-automated-human-activity-recognition
Repo
Framework

Loss is its own Reward: Self-Supervision for Reinforcement Learning


Title	Loss is its own Reward: Self-Supervision for Reinforcement Learning
Authors	Evan Shelhamer, Parsa Mahmoudieh, Max Argus, Trevor Darrell
Abstract	Reinforcement learning optimizes policies for expected cumulative reward. Need the supervision be so narrow? Reward is delayed and sparse for many tasks, making it a difficult and impoverished signal for end-to-end optimization. To augment reward, we consider a range of self-supervised tasks that incorporate states, actions, and successors to provide auxiliary losses. These losses offer ubiquitous and instantaneous supervision for representation learning even in the absence of reward. While current results show that learning from reward alone is feasible, pure reinforcement learning methods are constrained by computational and data efficiency issues that can be remedied by auxiliary losses. Self-supervised pre-training and joint optimization improve the data efficiency and policy returns of end-to-end reinforcement learning.
Tasks	Representation Learning
Published	2016-12-21
URL	http://arxiv.org/abs/1612.07307v2
PDF	http://arxiv.org/pdf/1612.07307v2.pdf
PWC	https://paperswithcode.com/paper/loss-is-its-own-reward-self-supervision-for
Repo
Framework

Corporate Disruption in the Science of Machine Learning


Title	Corporate Disruption in the Science of Machine Learning
Authors	Sam Work
Abstract	This MSc dissertation considers the effects of the current corporate interest on researchers in the field of machine learning. Situated within the field’s cyclical history of academic, public and corporate interest, this dissertation investigates how current researchers view recent developments and negotiate their own research practices within an environment of increased commercial interest and funding. The original research consists of in-depth interviews with 12 machine learning researchers working in both academia and industry. Building on theory from science, technology and society studies, this dissertation problematizes the traditional narratives of the neoliberalization of academic research by allowing the researchers themselves to discuss how their career choices, working environments and interactions with others in the field have been affected by the reinvigorated corporate interest of recent years.
Tasks
Published	2016-12-13
URL	http://arxiv.org/abs/1612.04108v1
PDF	http://arxiv.org/pdf/1612.04108v1.pdf
PWC	https://paperswithcode.com/paper/corporate-disruption-in-the-science-of
Repo
Framework

Modular Deep Q Networks for Sim-to-real Transfer of Visuo-motor Policies


Title	Modular Deep Q Networks for Sim-to-real Transfer of Visuo-motor Policies
Authors	Fangyi Zhang, Jürgen Leitner, Michael Milford, Peter Corke
Abstract	While deep learning has had significant successes in computer vision thanks to the abundance of visual data, collecting sufficiently large real-world datasets for robot learning can be costly. To increase the practicality of these techniques on real robots, we propose a modular deep reinforcement learning method capable of transferring models trained in simulation to a real-world robotic task. We introduce a bottleneck between perception and control, enabling the networks to be trained independently, but then merged and fine-tuned in an end-to-end manner to further improve hand-eye coordination. On a canonical, planar visually-guided robot reaching task a fine-tuned accuracy of 1.6 pixels is achieved, a significant improvement over naive transfer (17.5 pixels), showing the potential for more complicated and broader applications. Our method provides a technique for more efficient learning and transfer of visuo-motor policies for real robotic systems without relying entirely on large real-world robot datasets.
Tasks
Published	2016-10-21
URL	http://arxiv.org/abs/1610.06781v4
PDF	http://arxiv.org/pdf/1610.06781v4.pdf
PWC	https://paperswithcode.com/paper/modular-deep-q-networks-for-sim-to-real
Repo
Framework

Review of Fall Detection Techniques: A Data Availability Perspective


Title	Review of Fall Detection Techniques: A Data Availability Perspective
Authors	Shehroz S. Khan, Jesse Hoey
Abstract	A fall is an abnormal activity that occurs rarely; however, missing to identify falls can have serious health and safety implications on an individual. Due to the rarity of occurrence of falls, there may be insufficient or no training data available for them. Therefore, standard supervised machine learning methods may not be directly applied to handle this problem. In this paper, we present a taxonomy for the study of fall detection from the perspective of availability of fall data. The proposed taxonomy is independent of the type of sensors used and specific feature extraction/selection methods. The taxonomy identifies different categories of classification methods for the study of fall detection based on the availability of their data during training the classifiers. Then, we present a comprehensive literature review within those categories and identify the approach of treating a fall as an abnormal activity to be a plausible research direction. We conclude our paper by discussing several open research problems in the field and pointers for future research.
Tasks
Published	2016-05-30
URL	http://arxiv.org/abs/1605.09351v2
PDF	http://arxiv.org/pdf/1605.09351v2.pdf
PWC	https://paperswithcode.com/paper/review-of-fall-detection-techniques-a-data
Repo
Framework

Extended Graded Modalities in Strategy Logic


Title	Extended Graded Modalities in Strategy Logic
Authors	Benjamin Aminof, Vadim Malvone, Aniello Murano, Sasha Rubin
Abstract	Strategy Logic (SL) is a logical formalism for strategic reasoning in multi-agent systems. Its main feature is that it has variables for strategies that are associated to specific agents with a binding operator. We introduce Graded Strategy Logic (GradedSL), an extension of SL by graded quantifiers over tuples of strategy variables, i.e., “there exist at least g different tuples (x_1,…,x_n) of strategies” where g is a cardinal from the set N union {aleph_0, aleph_1, 2^aleph_0}. We prove that the model-checking problem of GradedSL is decidable. We then turn to the complexity of fragments of GradedSL. When the g’s are restricted to finite cardinals, written GradedNSL, the complexity of model-checking is no harder than for SL, i.e., it is non-elementary in the quantifier rank. We illustrate our formalism by showing how to count the number of different strategy profiles that are Nash equilibria (NE), or subgame-perfect equilibria (SPE). By analyzing the structure of the specific formulas involved, we conclude that the important problems of checking for the existence of a unique NE or SPE can both be solved in 2ExpTime, which is not harder than merely checking for the existence of such equilibria.
Tasks
Published	2016-07-12
URL	http://arxiv.org/abs/1607.03354v1
PDF	http://arxiv.org/pdf/1607.03354v1.pdf
PWC	https://paperswithcode.com/paper/extended-graded-modalities-in-strategy-logic
Repo
Framework

Discovery of Latent Factors in High-dimensional Data Using Tensor Methods


Title	Discovery of Latent Factors in High-dimensional Data Using Tensor Methods
Authors	Furong Huang
Abstract	Unsupervised learning aims at the discovery of hidden structure that drives the observations in the real world. It is essential for success in modern machine learning. Latent variable models are versatile in unsupervised learning and have applications in almost every domain. Training latent variable models is challenging due to the non-convexity of the likelihood objective. An alternative method is based on the spectral decomposition of low order moment tensors. This versatile framework is guaranteed to estimate the correct model consistently. My thesis spans both theoretical analysis of tensor decomposition framework and practical implementation of various applications. This thesis presents theoretical results on convergence to globally optimal solution of tensor decomposition using the stochastic gradient descent, despite non-convexity of the objective. This is the first work that gives global convergence guarantees for the stochastic gradient descent on non-convex functions with exponentially many local minima and saddle points. This thesis also presents large-scale deployment of spectral methods carried out on various platforms. Dimensionality reduction techniques such as random projection are incorporated for a highly parallel and scalable tensor decomposition algorithm. We obtain a gain in both accuracies and in running times by several orders of magnitude compared to the state-of-art variational methods. To solve real world problems, more advanced models and learning algorithms are proposed. This thesis discusses generalization of LDA model to mixed membership stochastic block model for learning user communities in social network, convolutional dictionary model for learning word-sequence embeddings, hierarchical tensor decomposition and latent tree structure model for learning disease hierarchy, and spatial point process mixture model for detecting cell types in neuroscience.
Tasks	Dimensionality Reduction, Latent Variable Models
Published	2016-06-10
URL	http://arxiv.org/abs/1606.03212v1
PDF	http://arxiv.org/pdf/1606.03212v1.pdf
PWC	https://paperswithcode.com/paper/discovery-of-latent-factors-in-high
Repo
Framework

Adaptive Filter for Automatic Identification of Multiple Faults in a Noisy OTDR Profile


Title	Adaptive Filter for Automatic Identification of Multiple Faults in a Noisy OTDR Profile
Authors	Jean Pierre von der Weid, Mario H. Souto, Joaquim D. Garcia, Gustavo C. Amaral
Abstract	We present a novel methodology able to distinguish meaningful level shifts from typical signal fluctuations. A two-stage regularization filtering can accurately identify the location of the significant level-shifts with an efficient parameter-free algorithm. The developed methodology demands low computational effort and can easily be embedded in a dedicated processing unit. Our case studies compare the new methodology with current available ones and show that it is the most adequate technique for fast detection of multiple unknown level-shifts in a noisy OTDR profile.
Tasks
Published	2016-02-13
URL	https://arxiv.org/abs/1602.04379v1
PDF	https://arxiv.org/pdf/1602.04379v1.pdf
PWC	https://paperswithcode.com/paper/adaptive-filter-for-automatic-identification
Repo
Framework

Deep Action Sequence Learning for Causal Shape Transformation


Title	Deep Action Sequence Learning for Causal Shape Transformation
Authors	Kin Gwn Lore, Daniel Stoecklein, Michael Davies, Baskar Ganapathysubramanian, Soumik Sarkar
Abstract	Deep learning became the method of choice in recent year for solving a wide variety of predictive analytics tasks. For sequence prediction, recurrent neural networks (RNN) are often the go-to architecture for exploiting sequential information where the output is dependent on previous computation. However, the dependencies of the computation lie in the latent domain which may not be suitable for certain applications involving the prediction of a step-wise transformation sequence that is dependent on the previous computation only in the visible domain. We propose that a hybrid architecture of convolution neural networks (CNN) and stacked autoencoders (SAE) is sufficient to learn a sequence of actions that nonlinearly transforms an input shape or distribution into a target shape or distribution with the same support. While such a framework can be useful in a variety of problems such as robotic path planning, sequential decision-making in games, and identifying material processing pathways to achieve desired microstructures, the application of the framework is exemplified by the control of fluid deformations in a microfluidic channel by deliberately placing a sequence of pillars. Learning of a multistep topological transform has significant implications for rapid advances in material science and biomedical applications.
Tasks	Decision Making
Published	2016-05-17
URL	http://arxiv.org/abs/1605.05368v3
PDF	http://arxiv.org/pdf/1605.05368v3.pdf
PWC	https://paperswithcode.com/paper/deep-action-sequence-learning-for-causal
Repo
Framework

A Tractable Fully Bayesian Method for the Stochastic Block Model


Title	A Tractable Fully Bayesian Method for the Stochastic Block Model
Authors	Kohei Hayashi, Takuya Konishi, Tatsuro Kawamoto
Abstract	The stochastic block model (SBM) is a generative model revealing macroscopic structures in graphs. Bayesian methods are used for (i) cluster assignment inference and (ii) model selection for the number of clusters. In this paper, we study the behavior of Bayesian inference in the SBM in the large sample limit. Combining variational approximation and Laplace’s method, a consistent criterion of the fully marginalized log-likelihood is established. Based on that, we derive a tractable algorithm that solves tasks (i) and (ii) concurrently, obviating the need for an outer loop to check all model candidates. Our empirical and theoretical results demonstrate that our method is scalable in computation, accurate in approximation, and concise in model selection.
Tasks	Bayesian Inference, Model Selection
Published	2016-02-06
URL	http://arxiv.org/abs/1602.02256v1
PDF	http://arxiv.org/pdf/1602.02256v1.pdf
PWC	https://paperswithcode.com/paper/a-tractable-fully-bayesian-method-for-the
Repo
Framework

On Spectral Analysis of Directed Signed Graphs


Title	On Spectral Analysis of Directed Signed Graphs
Authors	Yuemeng Li, Xintao Wu, Aidong Lu
Abstract	It has been shown that the adjacency eigenspace of a network contains key information of its underlying structure. However, there has been no study on spectral analysis of the adjacency matrices of directed signed graphs. In this paper, we derive theoretical approximations of spectral projections from such directed signed networks using matrix perturbation theory. We use the derived theoretical results to study the influences of negative intra cluster and inter cluster directed edges on node spectral projections. We then develop a spectral clustering based graph partition algorithm, SC-DSG, and conduct evaluations on both synthetic and real datasets. Both theoretical analysis and empirical evaluation demonstrate the effectiveness of the proposed algorithm.
Tasks
Published	2016-12-23
URL	http://arxiv.org/abs/1612.08102v1
PDF	http://arxiv.org/pdf/1612.08102v1.pdf
PWC	https://paperswithcode.com/paper/on-spectral-analysis-of-directed-signed
Repo
Framework

3D Face Reconstruction by Learning from Synthetic Data


Title	3D Face Reconstruction by Learning from Synthetic Data
Authors	Elad Richardson, Matan Sela, Ron Kimmel
Abstract	Fast and robust three-dimensional reconstruction of facial geometric structure from a single image is a challenging task with numerous applications. Here, we introduce a learning-based approach for reconstructing a three-dimensional face from a single image. Recent face recovery methods rely on accurate localization of key characteristic points. In contrast, the proposed approach is based on a Convolutional-Neural-Network (CNN) which extracts the face geometry directly from its image. Although such deep architectures outperform other models in complex computer vision problems, training them properly requires a large dataset of annotated examples. In the case of three-dimensional faces, currently, there are no large volume data sets, while acquiring such big-data is a tedious task. As an alternative, we propose to generate random, yet nearly photo-realistic, facial images for which the geometric form is known. The suggested model successfully recovers facial shapes from real images, even for faces with extreme expressions and under various lighting conditions.
Tasks	3D Face Reconstruction, Face Reconstruction
Published	2016-09-14
URL	http://arxiv.org/abs/1609.04387v2
PDF	http://arxiv.org/pdf/1609.04387v2.pdf
PWC	https://paperswithcode.com/paper/3d-face-reconstruction-by-learning-from
Repo
Framework

Unsupervised Category Discovery via Looped Deep Pseudo-Task Optimization Using a Large Scale Radiology Image Database


Title	Unsupervised Category Discovery via Looped Deep Pseudo-Task Optimization Using a Large Scale Radiology Image Database
Authors	Xiaosong Wang, Le Lu, Hoo-chang Shin, Lauren Kim, Isabella Nogues, Jianhua Yao, Ronald Summers
Abstract	Obtaining semantic labels on a large scale radiology image database (215,786 key images from 61,845 unique patients) is a prerequisite yet bottleneck to train highly effective deep convolutional neural network (CNN) models for image recognition. Nevertheless, conventional methods for collecting image labels (e.g., Google search followed by crowd-sourcing) are not applicable due to the formidable difficulties of medical annotation tasks for those who are not clinically trained. This type of image labeling task remains non-trivial even for radiologists due to uncertainty and possible drastic inter-observer variation or inconsistency. In this paper, we present a looped deep pseudo-task optimization procedure for automatic category discovery of visually coherent and clinically semantic (concept) clusters. Our system can be initialized by domain-specific (CNN trained on radiology images and text report derived labels) or generic (ImageNet based) CNN models. Afterwards, a sequence of pseudo-tasks are exploited by the looped deep image feature clustering (to refine image labels) and deep CNN training/classification using new labels (to obtain more task representative deep features). Our method is conceptually simple and based on the hypothesized “convergence” of better labels leading to better trained CNN models which in turn feed more effective deep image features to facilitate more meaningful clustering/labels. We have empirically validated the convergence and demonstrated promising quantitative and qualitative results. Category labels of significantly higher quality than those in previous work are discovered. This allows for further investigation of the hierarchical semantic nature of the given large-scale radiology image database.
Tasks
Published	2016-03-25
URL	http://arxiv.org/abs/1603.07965v1
PDF	http://arxiv.org/pdf/1603.07965v1.pdf
PWC	https://paperswithcode.com/paper/unsupervised-category-discovery-via-looped
Repo
Framework

Conversion of Artificial Recurrent Neural Networks to Spiking Neural Networks for Low-power Neuromorphic Hardware


Title	Conversion of Artificial Recurrent Neural Networks to Spiking Neural Networks for Low-power Neuromorphic Hardware
Authors	Peter U. Diehl, Guido Zarrella, Andrew Cassidy, Bruno U. Pedroni, Emre Neftci
Abstract	In recent years the field of neuromorphic low-power systems that consume orders of magnitude less power gained significant momentum. However, their wider use is still hindered by the lack of algorithms that can harness the strengths of such architectures. While neuromorphic adaptations of representation learning algorithms are now emerging, efficient processing of temporal sequences or variable length-inputs remain difficult. Recurrent neural networks (RNN) are widely used in machine learning to solve a variety of sequence learning tasks. In this work we present a train-and-constrain methodology that enables the mapping of machine learned (Elman) RNNs on a substrate of spiking neurons, while being compatible with the capabilities of current and near-future neuromorphic systems. This “train-and-constrain” method consists of first training RNNs using backpropagation through time, then discretizing the weights and finally converting them to spiking RNNs by matching the responses of artificial neurons with those of the spiking neurons. We demonstrate our approach by mapping a natural language processing task (question classification), where we demonstrate the entire mapping process of the recurrent layer of the network on IBM’s Neurosynaptic System “TrueNorth”, a spike-based digital neuromorphic hardware architecture. TrueNorth imposes specific constraints on connectivity, neural and synaptic parameters. To satisfy these constraints, it was necessary to discretize the synaptic weights and neural activities to 16 levels, and to limit fan-in to 64 inputs. We find that short synaptic delays are sufficient to implement the dynamical (temporal) aspect of the RNN in the question classification task. The hardware-constrained model achieved 74% accuracy in question classification while using less than 0.025% of the cores on one TrueNorth chip, resulting in an estimated power consumption of ~17 uW.
Tasks	Representation Learning
Published	2016-01-16
URL	http://arxiv.org/abs/1601.04187v1
PDF	http://arxiv.org/pdf/1601.04187v1.pdf
PWC	https://paperswithcode.com/paper/conversion-of-artificial-recurrent-neural
Repo
Framework

Distilling an Ensemble of Greedy Dependency Parsers into One MST Parser


Title	Distilling an Ensemble of Greedy Dependency Parsers into One MST Parser
Authors	Adhiguna Kuncoro, Miguel Ballesteros, Lingpeng Kong, Chris Dyer, Noah A. Smith
Abstract	We introduce two first-order graph-based dependency parsers achieving a new state of the art. The first is a consensus parser built from an ensemble of independently trained greedy LSTM transition-based parsers with different random initializations. We cast this approach as minimum Bayes risk decoding (under the Hamming cost) and argue that weaker consensus within the ensemble is a useful signal of difficulty or ambiguity. The second parser is a “distillation” of the ensemble into a single model. We train the distillation parser using a structured hinge loss objective with a novel cost that incorporates ensemble uncertainty estimates for each possible attachment, thereby avoiding the intractable cross-entropy computations required by applying standard distillation objectives to problems with structured outputs. The first-order distillation parser matches or surpasses the state of the art on English, Chinese, and German.
Tasks	Dependency Parsing
Published	2016-09-24
URL	http://arxiv.org/abs/1609.07561v1
PDF	http://arxiv.org/pdf/1609.07561v1.pdf
PWC	https://paperswithcode.com/paper/distilling-an-ensemble-of-greedy-dependency
Repo
Framework