Summary of From Pixels to Planning: Scale-free Active Inference, by Karl Friston et al.
From pixels to planning: scale-free active inference
by Karl Friston, Conor Heins, Tim Verbelen, Lancelot Da Costa, Tommaso Salvatori, Dimitrije Markovic, Alexander Tschantz, Magnus Koudahl, Christopher Buckley, Thomas Parr
First submitted to arxiv on: 27 Jul 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Neurons and Cognition (q-bio.NC)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary A novel discrete state-space model for generative modeling is proposed, which generalizes partially observed Markov decision processes by incorporating paths as latent variables, enabling active inference and learning in dynamic settings. The model, termed renormalising generative models (RGM), can be viewed as a discrete analogue of deep convolutional neural networks or continuous state-space models in generalized coordinates of motion. This scale-invariant model allows for the learning of compositionality over space and time, resulting in models of paths or orbits, i.e., events of increasing temporal depth and itinerancy. The paper demonstrates the automatic discovery, learning, and deployment of RGMs through a series of applications, including image classification, movie and music compression and generation, and Atari-like game learning. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper creates a new way to generate things, like images or movies, by using paths as hidden variables. It’s based on old ideas from computer science and physics, but makes them work together in a new way. The model can learn patterns and rules, and apply them to new situations. The authors show how this works by testing it with different tasks, like recognizing pictures, making music, or playing games. |
Keywords
» Artificial intelligence » Image classification » Inference
