The Wisdom of the Body Revisited
summary
Argues that many non‑neural networks implement proportional–integral control, modeling blood calcium regulation around ~9 mg/dL as a PI loop and proving that an internal integrator is necessary and sufficient for perfect adaptation in linear systems. Uses control‑theory, convolutional impulse‑response models and optimal control to analyze training and addiction: the generalized Bannister model explains progressive overload and a flipped opponent process yields a greedy tapering policy that minimizes cumulative dose and provably keeps outcomes above a safety floor.
AI-written summary · about
chapters (9)
- 0:00Introduction
- 1:09Body wisdom and homeostasis
- 9:06Calcium homeostasis models
- 17:46Perfect adaptation and control
- 25:24Adaptation syndrome and overcompensation
- 30:11Progressive overload training
- 36:14Addiction models and tapering
- 41:15Future therapy directions
- 42:43Closing and Q&A
related talks
51mThe Brain as an Efficient, Robust Adaptive Learner
19mDeep Reinforcement Learning based Insulin Controller for Effective Type-1 Diabetic Care
32mHow Neuroscience Makes AI Intelligent
01h 13mComputation in Networks of Neurons in the Brain II
41mArtificial Intelligence: Let's Think About Learning
49mAn Algorithmic Theory of Brain Networks
32mData-Driven Inverse Modeling with Incomplete Observations
35mTowards Biologically Plausible Deep Learning: ...