From a presenter's perspective the interrupting questions may have been mildly frustrating. From a viewer's perspective they were indispensable. I had exactly the same questions and the answers really helped me to understand the choices at a deeper level. Thank you!
@adibkarimi11339 ай бұрын
Great talk! It's super exciting how RL with transformers performs so good.
@prof_shixo10 ай бұрын
For the question at timepoint 20:20 regarding the Markovian property and the fact that we are shifting the state using positional embedding, I think that the presenter reply was not that accurate. While he referred to it as state, actually what is there is an observation instead (e.g., an RGB frame) while the actual state is the encoder embedding given the encoded observation sequence along other token types. That view makes the state representation still to be a Markovian one as the encoder's embedding (the true state) is encapsulating all the history in the sequence and it is consistent (i.e., gives the same state given the same sequence), once the encoder is trained and frozen. In brief, I guess that was a confusion between observations and states in the RL paradigm.
@sh4ny14 ай бұрын
Great observation !
@rtBen2 жыл бұрын
On sequence modelling: - talk dated in the future oct 11, 2022 - streamed in July 2022 - “to appear in” neurIPS 2021
@AdityaGrover2 жыл бұрын
ha thanks for pointing the typo! The talk was in Oct 2021.
@empi_ai2 жыл бұрын
brilliant talk! very clear and interesting
@albertwang59742 жыл бұрын
This talking illustrate what's time travelling, what a future talking!
@djethereal992 жыл бұрын
Great talk, really interesting work!
@LucasOSouza2 жыл бұрын
nice talk! got a bit spooked watching this today and seeing the date in the first slide 😅
@m.h.w2852 жыл бұрын
How time flies!
@jamesnatchwey19612 жыл бұрын
🤣🤗
@productlog58952 жыл бұрын
beautiful lecture
@imolafodor46678 ай бұрын
hi, how is this method considered to be offline rl if at some point you are calling env.step() aka an online step?
@miquelnogueralonso25762 жыл бұрын
Are the slides available ?
@amrahmed20092 жыл бұрын
Are the slides available, please?
@markcarter6333 Жыл бұрын
Good stuff. Would it be better to leave questions in the chat? Or at the end of talk?
@zyzhang1130 Жыл бұрын
Are LLMs stable to train though
@sdfafds6823 Жыл бұрын
at the third page, "Larger model require fewer samples to reach the same performance ." Isn't that a little bit anti common sense? In my original opinion, more complex model needs to ingest more data.