The submission portal for the TTIC Adaptive Learning workshop is still open!

We're organizing a workshop! It's going to be a lot of fun, please do submit your work and attend :)

REMINDER: TTIC will be hosting a workshop on Adaptive Learning in Complex Environments July 9-11, organized by Abhishek Gupta, Eric Mazumdar, Aldo Pacchiano, and co-sponsored by IDEAL. Register for free: buff.ly/3JM5TgG

We had an amazing workshop!

We are organizing the deployableRL workshop at RLC. We’ll have a panel session with great panelists. We have prepared a questionnaire to gather topics folks would be interested in discussing. Let us know your thoughts! Deployable RL Workshop Yonathan Efroni sites.google.com/bu.edu/deploya…

We are organizing the deployableRL workshop at RLC. We’ll have a panel session with great panelists. We have prepared a questionnaire to gather topics folks would be interested in discussing. Let us know your thoughts! Deployable RL Workshop Yonathan Efroni sites.google.com/bu.edu/deploya…

Nice!

Deployable RL Workshop workshop is happening this Friday at RL_Conference. I will be moderating a panel hosting Minmin Chen, Omer Gottsman, Daniel Russo, Kaushik Subramanian and Cathy Wu. Please vote here docs.google.com/forms/d/e/1FAI… to push your favorite production RL question to the top!

I am @ RLC today! BU student Aida Afshar is presenting our work on how to efficiently adapt to the best learning rate parameter for any RL algorithm using online model selection techniques at the ICBINB workshop. Preprint below! arxiv.org/abs/2408.04046

Great keynote by David Silver, arguing that we need to re-focus on RL to get out of the LLM Valley RL_Conference

We are doing a post-covid restart of the NYC ML Symposium (events.nyas.org/event/ml2024/s… ) on Oct 18th. It's a great chance to meet folks in the area, so drop by if interested. Abstracts for posters are due Sept 6th, in about 10 days. (These are super quick & easy.)

There are so many interesting criticisms of LLMs as tools, wish folks would pick something beside tokenizer bugs: 1) is training on the output of a black box sufficient to constrain the internals? 2) is RlHF enough to teach a behavior or just something that looks like it

I agree!