Model Based Deep Reinforcement Learning: Understanding DreamerV3
Details
In past events, we have discussed how the interplay of the basal ganglia and neocortex in the mammalian brain may be implementing Model Based Reinforcement Learning.This enables mammals to learn more complex relationships than other cold blooded animals and increases the efficiency of learning too. Now we want to study the Deep Learning analog of this idea.
In this event, we will discuss DreamerV3. This is a SOTA Deep Learning algorithm for Model Based Reinforcement Learning. You can find the paper here: Mastering Diverse Domains through World Models
This general algorithm outperforms specialised methods in over 150 diverse tasks, and can solve challenging problems without extensive exploration, making the algorithm widely applicable.
In short, exciting stuff!
We will meet around 18:30. We will spend the first half an hour socializing with pizza and drinks. Then, from 19:00 onwards, we will start discussing.
The end time is open ended. We have put 21:30 tentatively, but the space allows us to discuss and network longer if we wish to.
There are no speakers in this meetup. We sit around a table as equals and discuss/brainstorm together. Please try to read the paper before joining, so that our discussion can go deep.
Looking forward to seeing you there!
Additional Resources
- DreamerV1 (first version of the algorithm published in 2020; easier to understand than the V3 version and contains lot more context).
- World Models (Earlier work in 2018 containing many of the core ideas and featuring verbose explanations)
- JAX Implementation from the first author
- Project Website
- Implementation in Ray RLLib