DeepAI AI Chat
Log In Sign Up

Toward an AI Physicist for Unsupervised Learning

by   Tailin Wu, et al.

We investigate opportunities and challenges for improving unsupervised machine learning using four common strategies with a long history in physics: divide-and-conquer, Occam's Razor, unification, and lifelong learning. Instead of using one model to learn everything, we propose a novel paradigm centered around the learning and manipulation of *theories*, which parsimoniously predict both aspects of the future (from past observations) and the domain in which these predictions are accurate. Specifically, we propose a novel generalized-mean-loss to encourage each theory to specialize in its comparatively advantageous domain, and a differentiable description length objective to downweight bad data and "snap" learned theories into simple symbolic formulas. Theories are stored in a "theory hub", which continuously unifies learned theories and can propose theories when encountering new environments. We test our implementation, the "AI Physicist" learning agent, on a suite of increasingly complex physics environments. From unsupervised observation of trajectories through worlds involving random combinations of gravity, electromagnetism, harmonic motion and elastic bounces, our agent typically learns faster and produces mean-squared prediction errors about a billion times smaller than a standard feedforward neural net of comparable complexity, typically recovering integer and rational theory parameters exactly. Our agent successfully identifies domains with different laws of motion also for a nonlinear chaotic double pendulum in a piecewise constant force field.


page 2

page 6


From proof theory to theories theory

In the last decades, several objects such as grammars, economical agents...

A probabilistic theory of trust concerning artificial intelligence: can intelligent robots trust humans?

In this paper, I argue for a probabilistic theory of trust, and the plau...

ToyArchitecture: Unsupervised Learning of Interpretable Models of the World

Research in Artificial Intelligence (AI) has focused mostly on two extre...

Rademacher complexity and spin glasses: A link between the replica and statistical theories of learning

Statistical learning theory provides bounds of the generalization gap, u...

How to select an objective function using information theory

Science tests competing theories or models by evaluating the similarity ...

Accelerating the BSM interpretation of LHC data with machine learning

The interpretation of Large Hadron Collider (LHC) data in the framework ...

Machine Learning for Discovering Effective Interaction Kernels between Celestial Bodies from Ephemerides

Building accurate and predictive models of the underlying mechanisms of ...