Large text-to-video models trained on internet-scale data have demonstra...
A goal of artificial intelligence is to construct an agent that can solv...
Future- or return-conditioned supervised learning is an emerging paradig...
It is common to address the curse of dimensionality in Markov decision
p...
A longstanding goal of the field of AI is a strategy for compiling diver...
Imitation learning aims to extract high-performance policies from logged...
Conventionally, generation of natural language for dialogue agents may b...
Goal-oriented dialogue systems face a trade-off between fluent language
...
The aim in imitation learning is to learn effective policies by utilizin...
Transformers provide a class of expressive architectures that are extrem...
In imitation learning, it is common to learn a behavior policy to match ...
Off-policy evaluation (OPE) holds the promise of being able to leverage
...
Reliable automatic evaluation of dialogue systems under an interactive
e...
The recent success of supervised learning methods on ever larger offline...
The presence of uncertainty in policy evaluation significantly complicat...
The recently proposed distribution correction estimation (DICE) family o...
Recently there has been growing interest in modeling sets with
exchangea...
Interpretability is rising as an important area of research in machine
l...
The performance bottlenecks of graph applications depend not only on the...