Tom Bush (@_tom_bush) header
Tom Bush (@_tom_bush) profile picture

AI DPhil-ing @ Oxford

Joined November 2023
Photo by Tom Bush (@_tom_bush)Photo by Tom Bush (@_tom_bush)Photo by Tom Bush (@_tom_bush)Photo by Tom Bush (@_tom_bush)Photo by Tom Bush (@_tom_bush)Photo by Tom Bush (@_tom_bush)Photo by Tom Bush (@_tom_bush)Photo by Tom Bush (@_tom_bush)Photo by Tom Bush (@_tom_bush)Photo by Tom Bush (@_tom_bush)
🤖 !! Model-free agents can internally plan !! 🤖 In our ICLR 2025 paper, we interpret a model-free RL agent and show that it internally performs a form of planning resembling bidirectional search.
7
46
7
316
38,857
Attending #ICLR2025 to present this research - please reach out if you want to chat about interp, reasoning, or anything else!!
🤖 Model-free agents can internally plan! Sokoban agents develop bidirectional search planning! 🔬 We probe for planning concepts ⚙️ Investigate plan formation ✅ Verify plans impact behavior Chat with us at #ICLR2025! 📍 Apr 24: Poster 10am + Oral 4:06pm SGT
2
17
1,019
Preprint time: “AI in a vat: Fundamental limits of efficient world modelling for agent sandboxing and interpretability” arxiv.org/abs/2504.04608 Exploring the fundamental limits that shape the design space of world modelling for agent sandboxing and interpretability
3
18
3
57
4,599
For instance, in the example below, the agent initially plans to push the upper-left box to the center-most by pushing it down and then right. However, the agent realises this is infeasible – once the box is been pushed down, it is stuck – and then forms an alternate plan.
1
6
843