Screening brief
What this video covers
John Schulman, identified in the catalogue as an OpenAI cofounder, gives an EECS colloquium on reinforcement learning from human feedback (RLHF). The talk reviews progress made using preference-based training, outlines practical limitations encountered in RLHF workflows, and discusses trade-offs relevant to aligning model behavior with human preferences.
The event is listed as an EECS Colloquium held on April 19, 2023 at Banatao Auditorium. The catalogue entry records the talk as published three years ago on the UC Berkeley EECS channel, with a duration of 1:03:32 and tagged under RLHF.
Summary and topic guide by HumanData.TV, based on the original publisher’s description and our editorial catalogue. This is not a transcript or independent verification of the speaker’s claims.
