One Policy to Control Them All: Shared Modular Policies for Agent-Agnostic Control

ICML 2020 huangwl18/modular-rl

We observe that a wide variety of drastically diverse locomotion styles across morphologies as well as centralized coordination emerges via message passing between decentralized modules purely from the reinforcement learning objective.

52
13 Jul 2020

Lifted Disjoint Paths with Application in Multiple Object Tracking

ICML 2020 AndreaHor/LifT_Solver

We present an extension to the disjoint paths problem in which additional \emph{lifted} edges are introduced to provide path connectivity priors.

MULTI-OBJECT TRACKING MULTIPLE OBJECT TRACKING

9
12 Jul 2020

WormPose: Image synthesis and convolutional networks for pose estimation in C. elegans

bioRxiv 2020 iteal/wormpose

An important model system for understanding genes, neurons and behavior, the nematode worm C. elegans naturally moves through a variety of complex postures, for which estimation from video data is challenging.

ANIMAL POSE ESTIMATION IMAGE GENERATION

4
10 Jul 2020

SUNRISE: A Simple Unified Framework for Ensemble Learning in Deep Reinforcement Learning

9 Jul 2020pokaxpoka/sunrise

SUNRISE integrates three key ingredients: (a) bootstrap with random initialization which improves the stability of the learning process by training a diverse ensemble of agents, (b) weighted Bellman backups, which prevent error propagation in Q-learning by reweighing sample transitions based on uncertainty estimates from the ensembles, and (c) an inference method that selects actions using highest upper-confidence bounds for efficient exploration.

EFFICIENT EXPLORATION Q-LEARNING

27
09 Jul 2020

IALE: Imitating Active Learner Ensembles

9 Jul 2020crispchris/IALE

As the performance of well-known AL heuristics highly depends on the underlying model and data, recent heuristic-independent approaches that are based on reinforcement learning directly learn a policy that makes use of the labeling history to select the next sample.

ACTIVE LEARNING IMITATION LEARNING

2
09 Jul 2020

Interpolated corrected curvature measures for polygonal surfaces

Computer Graphics Forum (Proceedings of Symposium on Geometry Processing) 2020 dcoeurjo/CorrectedNormalCurrent

We pro- pose a new framework to define curvature measures, based on the Corrected Normal Current, which generalizes the normal cycle: it uncouples the positional information of the polyhedral mesh from its geometric normal vector field, and the user can freely choose the corrected normal vector field at vertices for curvature computations.

3
09 Jul 2020

Auxiliary Tasks Speed Up Learning PointGoal Navigation

9 Jul 2020joel99/habitat-pointnav-aux

PointGoal Navigation is an embodied task that requires agents to navigate to a specified point in an unseen environment.

POINTGOAL NAVIGATION

4
09 Jul 2020

Recurrent Neural-Linear Posterior Sampling for Non-Stationary Contextual Bandits

9 Jul 2020paulorauber/rnlps

An agent in a non-stationary contextual bandit problem should balance between exploration and the exploitation of (periodic or structured) patterns present in its previous experiences.

MULTI-ARMED BANDITS

3
09 Jul 2020

Modelling the Distribution of 3D Brain MRI using a 2D Slice VAE

9 Jul 2020voanna/slices-to-3d-brain-vae

We propose a method to model 3D MR brain volumes distribution by combining a 2D slice VAE with a Gaussian model that captures the relationships between slices.

0
09 Jul 2020