LESSWRONG
LW

Khanh Nguyen

Message

In-coming postdoc at CHAI, UC-Berkeley, working under Prof. Stuart Russell's supervision

Khanh Nguyen has not written any posts yet.

Message

In-coming postdoc at CHAI, UC-Berkeley, working under Prof. Stuart Russell's supervision

1 karma

1 comment

Member for 3 years

Khanh Nguyen — LessWrong

Message

In-coming postdoc at CHAI, UC-Berkeley, working under Prof. Stuart Russell's supervision

Khanh Nguyen has not written any posts yet.

Message

In-coming postdoc at CHAI, UC-Berkeley, working under Prof. Stuart Russell's supervision

1 karma

1 comment

Member for 3 years

Nice work! I agree with many of the opinions. Nevertheless, the survey is still missing several key citations.

Personally, I have made several contributions to RLHF:

Reinforcement learning for bandit neural machine translation with simulated human feedback (Nguyen et al., 2017) is the first paper that shows the potential of using noisy rating feedback to train text generators.
Interactive learning from activity description (Nguyen et al., 2021) is one of the first frameworks for learning from descriptive language feedback with theoretical guarantees
Language models are bounded pragmatic speakers: Understanding RLHF from a Bayesian cognitive modeling perspective (Nguyen 2023) proposes a theoretical framework that makes it easy to understand the limitations of RLHF and derive improve directions.

Recently,... (read more)

-1