Why is AI hard and Physics simple?
arXiv:2104.00008
Abstract
We discuss why AI is hard and why physics is simple. We discuss how physical intuition and the approach of theoretical physics can be brought to bear on the field of artificial intelligence and specifically machine learning. We suggest that the underlying project of machine learning and the underlying project of physics are strongly coupled through the principle of sparsity, and we call upon theoretical physicists to work on AI as physicists. As a first step in that direction, we discuss an upcoming book on the principles of deep learning theory that attempts to realize this approach.
written for a special issue of Machine Learning: Science and Technology as an invited perspective piece
References in corpus (16)
- Distributed Representations of Words and Phrases and their Compositionality
- The World as a Hologram
- Language Models are Few-Shot Learners
- A bound on chaos
- Black holes and the butterfly effect
- Scaling Laws for Neural Language Models
- Black holes as mirrors: quantum information in random subsystems
- Neural Tangent Kernel: Convergence and Generalization in Neural Networks
- Dota 2 with Large Scale Deep Reinforcement Learning
- Complexity Equals Action
- Complexity, action, and black holes
- Towards the fast scrambling conjecture
- On Lazy Training in Differentiable Programming
- The Typical-State Paradox: Diagnosing Horizons with Complexity
- Scrambling speed of random quantum circuits
- TASI lectures on complex structures