As my first post on this platform, allow me to advertise the RL theory lecture notes I have been developing with Sasha Rakhlin: arxiv.org/abs/2312.16730 (shameless repost of my pinned tweet) - ThreadSky

About ThreadSky

djfoster.bsky.social • 98 days ago

As my first post on this platform, allow me to advertise the RL theory lecture notes I have been developing with Sasha Rakhlin: https://arxiv.org/abs/2312.16730

(shameless repost of my pinned tweet)

Comments

djfoster.bsky.social•98 days ago

These lecture notes are based on a class we taught at MIT in Fall '22 and '23 (https://mit.edu/~rakhlin/course-decision-making-f23.html).

djfoster.bsky.social•98 days ago

They aim to teach the foundations of reinforcement learning (and more broadly, interactive decision making) using the DEC (Decision-Estimation Coefficient) machinery we have been developing over the few years (e.g., https://arxiv.org/abs/2112.13487) to offer a unifying perspective.

djfoster.bsky.social•98 days ago

We begin from the simplest decision making problem, multi-armed bandits, then gradually add structure, building up to reinforcement learning, with connections and parallels to supervised learning/estimation as an overarching theme.

djfoster.bsky.social•98 days ago

The notes are a work in progress---please let us know if you have feedback!

robert-g-campbell.bsky.social•97 days ago

Hey Dylan, thanks for making these available! Looks like the link is broken though.

johnegan.bsky.social•95 days ago

hi dylan,

am trying to develop options for probabilistic firewalls
Q: what is/are the best security measure(s) that you are aware of to help stop or mitigate probabilistic injection ?
the simplest form of probabilistic injection is a ‘prompt injection’

abhishekshar.bsky.social•97 days ago

The notes are great! Thank you!

Posting Rules

Be respectful to others
No spam or self-promotion
Stay on topic
Follow Bluesky's terms of service

Comments

Posting Rules

Reply