Learned models of the environment provide reinforcement learning (RL) ag...
Off-policy learning allows us to learn about possible policies of behavi...
Supporting state-of-the-art AI research requires balancing rapid prototy...
We propose a novel policy update that combines regularized policy
optimi...
Temporal abstractions in the form of options have been shown to help
rei...
Reinforcement learning (RL) algorithms update an agent's parameters acco...
Deep reinforcement learning includes a broad family of algorithms that
p...
Reinforcement learning (RL) algorithms often require expensive manual or...
Reinforcement learning agents can include different components, such as
...
We investigate the combination of actor-critic reinforcement learning
al...
Arguably, intelligent agents ought to be able to discover their own ques...
This paper introduces the Behaviour Suite for Reinforcement Learning, or...
We consider a general class of non-linear Bellman equations. These open ...
Many deep reinforcement learning algorithms contain inductive biases tha...
We examine the question of when and how parametric models are most usefu...
The ability to transfer skills across tasks has the potential to scale u...
Currently the only techniques for sharing governance of a deep learning ...
We know from reinforcement learning theory that temporal difference lear...
The reinforcement learning community has made great strides in designing...
Despite significant advances in the field of deep Reinforcement Learning...
We propose a distributed architecture for deep reinforcement learning at...
Some real-world domains are best characterized as a single task, but for...
The deep reinforcement learning community has made several independent
i...
One of the key challenges of artificial intelligence is to learn models ...
Most learning algorithms are not invariant to the scale of the function ...