Publications related to Understanding Deep Neural Function Approximation in Reinforcement Learning via ϵ-Greedy Exploration

Entropy and mutual information in models of deep neural networks

Nicolas Macris, Florent Gérard Krzakala, Lenka Zdeborová, Jean François Emmanuel Barbier, Clément Dominique Luneau

We examine a class of stochastic deep learning models with a tractable method to compute information-theoretic quantities. Our contributions are three-fold: (i) We show how entropies and mutual informations can be derived from heuristic statistical physics ...

NEURAL INFORMATION PROCESSING SYSTEMS (NIPS)2018

Deep Learning Works in Practice. But Does it Work in Theory?

Rachid Guerraoui, Le Nguyen Hoang

Deep learning relies on a very specific kind of neural networks: those superposing several neural layers. In the last few years, deep learning achieved major breakthroughs in many tasks such as image analysis, speech recognition, natural language processin ...

2018

Learning Continuous Muscle Control for a Multi-joint Arm by Extending Proximal Policy Optimization with a Liquid State Machine

Marc-Oliver Gewaltig

There have been many advances in the field of reinforcement learning in continuous control problems. Usually, these approaches use deep learning with artificial neural networks for approximation of policies and value functions. In addition, there have been ...

SPRINGER INTERNATIONAL PUBLISHING AG2018

Robustness of Classifiers to Universal Perturbations: A Geometric Perspective

Pascal Frossard, Seyed Mohsen Moosavi Dezfooli, Alhussein Fawzi

Deep networks have recently been shown to be vulnerable to universal perturbations: there exist very small image-agnostic perturbations that cause most natural images to be misclassified by such classifiers. In this paper, we provide a quantitative analysi ...

2018

LIFT: Learned Invariant Feature Transform

Pascal Fua, Vincent Lepetit, Kwang Moo Yi, Eduard Trulls Fortuny

We introduce a novel Deep Network architecture that implements the full feature point handling pipeline, that is, detection, orientation estimation, and feature description. While previous works have successfully tackled each one of these problems individu ...

Springer Int Publishing Ag2016

Towards Viewpoint Invariant 3D Human Pose Estimation

Alexandre Massoud Alahi

We propose a viewpoint invariant model for 3D human pose estimation from a single depth image. To achieve this, our discriminative model embeds local regions into a learned viewpoint invariant feature space. Formulated as a multi-task learning problem, our ...

Springer2016

Well-Posedness, Regularity, and Convergence Analysis of the Finite Element Approximation of a Generalized Robin Boundary Value Problem

Alfio Quarteroni, Luca Dede', Claudia Maria Colciago

In this paper, we propose the mathematical and finite element analysis of a second-order partial differential equation endowed with a generalized Robin boundary condition which involves the Laplace--Beltrami operator by introducing a function space $H^1(\O ...

Society for Industrial and Applied Mathematics2015

On Neural Networks’ Ability to Approximate Geometrical Variation Propagation in Assembly

Loïc Andolfatto

Abstract Tolerance analysis is an important step to validate assembly process planning scenario. Simulations are generally performed to evaluate the expected geometrical variations of the assembled product. When the simulation models take into account part ...

2013

On Neural Networks’ Ability to Approximate Geometrical Variation Propagation in Assembly

Loïc Andolfatto

Tolerance analysis is an important step to validate assembly process planning scenario. Simulations are generally performed to evaluate the expected geometrical variations of the assembled product. When the simulation models take into account part complian ...

2013

Network Simplification: the Gaussian N-Relay Diamond Network

Christina Fragouli, Ayfer Özgür Aydin

We consider the Gaussian

N

-relay diamond network, where a source wants to communicate to a destination through a layer of

N

-relay nodes. We investigate the following question: What fraction of the capacity can we maintain by using only

k

out of the $ ...

2010

Understanding Deep Neural Function Approximation in Reinforcement Learning via ϵ-Greedy Exploration

Graph Chatbot

Chat with Graph Search

Entropy and mutual information in models of deep neural networks

Deep Learning Works in Practice. But Does it Work in Theory?

Learning Continuous Muscle Control for a Multi-joint Arm by Extending Proximal Policy Optimization with a Liquid State Machine

Robustness of Classifiers to Universal Perturbations: A Geometric Perspective

LIFT: Learned Invariant Feature Transform

Towards Viewpoint Invariant 3D Human Pose Estimation

Well-Posedness, Regularity, and Convergence Analysis of the Finite Element Approximation of a Generalized Robin Boundary Value Problem

On Neural Networks’ Ability to Approximate Geometrical Variation Propagation in Assembly

On Neural Networks’ Ability to Approximate Geometrical Variation Propagation in Assembly

Network Simplification: the Gaussian N-Relay Diamond Network

Entropy and mutual information in models of deep neural networks

Learning Continuous Muscle Control for a Multi-joint Arm by Extending Proximal Policy Optimization with a Liquid State Machine

LIFT: Learned Invariant Feature Transform

Deep Learning Works in Practice. But Does it Work in Theory?

Robustness of Classifiers to Universal Perturbations: A Geometric Perspective

Towards Viewpoint Invariant 3D Human Pose Estimation

On Neural Networks’ Ability to Approximate Geometrical Variation Propagation in Assembly

Well-Posedness, Regularity, and Convergence Analysis of the Finite Element Approximation of a Generalized Robin Boundary Value Problem

On Neural Networks’ Ability to Approximate Geometrical Variation Propagation in Assembly

Network Simplification: the Gaussian N-Relay Diamond Network