I adopted the concept of seeds from the book, the creative act by Rick Rubin here to document all interesting ideas that I want to pursue scientifically: (they will be organized into blog posts)
- Inductive bias
- -> Causality
- OOD generalization (related to Causality)
- Separating the world model from the inference machine
- How does GFlowNet achieve this?
- “Inference” here is an abstract concept. ML/DL is a method to achieve it but there are others.
- Modularity and reusability of knowledge
- Abstraction
- Uncertainty in world model
- -> Emergence and complexity
- Consciousness
- Multimodal distribution over inference results
- Stepping as a way to represent arbitrarily rich distributions
Most seeds above come from https://yoshuabengio.org/2023/03/21/scaling-in-the-service-of-reasoning-model-based-ml/
-> GFlowNet is a manifestation of all these ideas.
More seeds
- Counterfactuals
- Understanding and Explanations
Leave a comment