Comments
Hacker News
things that are hard to verify.
ai didnt takeover the world like it was supposed to because its only good at verifiable stuff like coding but vast majority of economy isnt verifiable.
by dominotw
On Policy Self Distillation and Active Learning. Anything that increases sample efficiency by providing a richer more dense feedback signal and is more efficient at exploration / sampling.
by m_ke
I would probably follow whatever Farama is doing, https://farama.org/projects especially any of the (MO) multi-objective stuff ...
by jonbaer
RL for closed-loop adaptive BCIs and sim-to-real robotics still look like the strongest practical directions right now.
Intrinsic motivation/curiosity driven exploration/novelty seeking feels like it has a breakthrough paper waiting. If someone could get those methods working for LLMs, we start getting things like move37 but in math proofs and then everything else
by itkovian_
My advice, as somebody who has a master’s degree, is to pick a topic that you are really interested in, and that captures your imagination. You will have to dedicate yourself to this work for a considerable amount of time.
Doing a CV-driven master thesis will be miserable and more likely to fail. It should be something that is fun and captivating.
by Hendrikto
General advice from a CV PhD dropout:
Lean onto the interests of assistant professors at your school instead of trying to find promising RL subtopics.
This way you can get time with someone who’s deep in the field and can be a guiding light for your projects.
If you’re really into embodied AI, you can add that to an existing project but doing it by yourself can be extremely tough especially if your topic requires a lot of compute and your last name is not Rockefeller.
by gessha
Join the discussion
Write your take first — we'll ask for email only when you're ready to publish.
- Hacker News
- things that are hard to verify.
ai didnt takeover the world like it was supposed to because its only good at verifiable stuff like coding but vast majority of economy isnt verifiable.
by dominotw - On Policy Self Distillation and Active Learning. Anything that increases sample efficiency by providing a richer more dense feedback signal and is more efficient at exploration / sampling.by m_ke
- I would probably follow whatever Farama is doing, https://farama.org/projects especially any of the (MO) multi-objective stuff ...by jonbaer
- RL for closed-loop adaptive BCIs and sim-to-real robotics still look like the strongest practical directions right now.
- Intrinsic motivation/curiosity driven exploration/novelty seeking feels like it has a breakthrough paper waiting. If someone could get those methods working for LLMs, we start getting things like move37 but in math proofs and then everything elseby itkovian_
- My advice, as somebody who has a master’s degree, is to pick a topic that you are really interested in, and that captures your imagination. You will have to dedicate yourself to this work for a considerable amount of time.
Doing a CV-driven master thesis will be miserable and more likely to fail. It should be something that is fun and captivating.
by Hendrikto - General advice from a CV PhD dropout:
Lean onto the interests of assistant professors at your school instead of trying to find promising RL subtopics.
This way you can get time with someone who’s deep in the field and can be a guiding light for your projects.
If you’re really into embodied AI, you can add that to an existing project but doing it by yourself can be extremely tough especially if your topic requires a lot of compute and your last name is not Rockefeller.
by gessha