A disentangled recognition and nonlinear dynamics model for unsupervised learning
M Fraccaro, S Kamronn, U Paquet… - Advances in neural …, 2017 - proceedings.neurips.cc
This paper takes a step towards temporal reasoning in a dynamically changing video, not in
the pixel space that constitutes its frames, but in a latent space that describes the non-linear …
the pixel space that constitutes its frames, but in a latent space that describes the non-linear …
Recurrent environment simulators
Models that can simulate how environments change in response to actions can be used by
agents to plan and act efficiently. We improve on previous environment simulators from high …
agents to plan and act efficiently. We improve on previous environment simulators from high …
Provably efficient reinforcement learning in partially observable dynamical systems
Abstract We study Reinforcement Learning for partially observable systems using function
approximation. We propose a new PO-bilinear framework, that is general enough to include …
approximation. We propose a new PO-bilinear framework, that is general enough to include …
Pac reinforcement learning for predictive state representations
In this paper we study online Reinforcement Learning (RL) in partially observable dynamical
systems. We focus on the Predictive State Representations (PSRs) model, which is an …
systems. We focus on the Predictive State Representations (PSRs) model, which is an …
First-person activity forecasting with online inverse reinforcement learning
N Rhinehart, KM Kitani - Proceedings of the IEEE …, 2017 - openaccess.thecvf.com
We address the problem of incrementally modeling and forecasting long-term goals of a first-
person camera wearer: what the user will do, where they will go, and what goal they seek. In …
person camera wearer: what the user will do, where they will go, and what goal they seek. In …
Provably efficient imitation learning from observation alone
Abstract We study Imitation Learning (IL) from Observations alone (ILFO) in large-scale
MDPs. While most IL algorithms rely on an expert to directly provide actions to the learner, in …
MDPs. While most IL algorithms rely on an expert to directly provide actions to the learner, in …
Truncated horizon policy search: Combining reinforcement learning & imitation learning
In this paper, we propose to combine imitation and reinforcement learning via the idea of
reward shaping using an oracle. We study the effectiveness of the near-optimal cost-to-go …
reward shaping using an oracle. We study the effectiveness of the near-optimal cost-to-go …
Dual policy iteration
Recently, a novel class of Approximate Policy Iteration (API) algorithms have demonstrated
impressive practical performance (eg, ExIt from [1], AlphaGo-Zero from [2]). This new family …
impressive practical performance (eg, ExIt from [1], AlphaGo-Zero from [2]). This new family …
Embed to control partially observed systems: Representation learning with provable sample efficiency
Reinforcement learning in partially observed Markov decision processes (POMDPs) faces
two challenges.(i) It often takes the full history to predict the future, which induces a sample …
two challenges.(i) It often takes the full history to predict the future, which induces a sample …
Pves: Position-velocity encoders for unsupervised learning of structured state representations
We propose position-velocity encoders (PVEs) which learn---without supervision---to
encode images to positions and velocities of task-relevant objects. PVEs encode a single …
encode images to positions and velocities of task-relevant objects. PVEs encode a single …