The Power of Collective Memory
summary
Describes the Amoebot model of anonymous lattice particles with only constant-size memory and local sensing, proves a Markov-chain “compression” algorithm where lambda > 2+√2 yields configurations within constant factor of minimum perimeter while smaller lambda (>1) provokes spindly maximized-perimeter shapes, and implements the memory-free embodiment on 2 cm BobBots using tunable magnets with matching physics simulations and microfluidic experiments indicating the predicted aggregation/dispersion phase change.
AI-written summary · about
chapters (8)
- 0:00Introduction
- 1:26Self-organization in nature
- 8:20Constant-memory algorithms
- 22:39Algorithms without memory
- 35:01From theory to robots
- 38:57Microscale robots
- 41:03Conclusion
- 42:30Collaborations
related talks
59mCal Day Lecture: Can a Machine be Conscious? Towards a Conscious AI
13mFabObscura: Computational Design and Fabrication for Interactive Barrier-Grid Animations
26mWhile(1) i++ — How far can it go?
13mMemPool Flavors: Between Versatility and Specialization in a RISC-V Manycore Cluster
06mWhen Collaboration Betrays: The Evolving Attack Surface of...
15mProgramming and Reasoning with Partial Observability
30mSymbolic Execution of Hadamard-Toffoli Quantum Circuits
27mScheme Pearl: Quantum Continuations