The explosive growth of language models and their applications have led ...
Sparse expert models are a thirty-year old concept re-emerging as a popu...
There have been a lot of interest in the scaling properties of Transform...
Scale has opened new frontiers in natural language processing – but at a...
Research on exploration in reinforcement learning, as applied to Atari 2...
There remain many open questions pertaining to the scaling behaviour of
...
Novel computer vision architectures monopolize the spotlight, but the im...
The research community has proposed copious modifications to the Transfo...
In deep learning, models typically reuse the same parameters for all inp...
Experience replay is central to off-policy algorithms in deep reinforcem...
Model-free deep reinforcement learning algorithms are troubled with poor...
Text-based games are a natural challenge domain for deep reinforcement
l...
This paper provides an empirical evaluation of recently developed explor...
Reinforcement learning (RL) typically defines a discount factor as part ...
Generating high-quality text with sufficient diversity is essential for ...
We present Deep Graph Infomax (DGI), a general approach for learning nod...
In many environments only a tiny subset of all states yield high reward....
It has been postulated that a good representation is one that disentangl...
Neural text generation models are often autoregressive language models o...
Generative adversarial networks (GANs) are a family of generative models...