(2016. 4) ML Learn And Think Like Human
Submitted on 2016. 4
Brenden M. Lake, Tomer D. Ullman, Joshua B. Tenenbaum and Samuel J. Gershman
Simple Summary
review progress in cognitive science suggesting that truly human-like learning and thinking machines will have to reach beyond current engineering trends in both what they learn, and how they learn it.
(a) build causal models of the world that support explanation and understanding, rather than merely solving pattern recognition problems; (b) ground learning in intuitive theories of physics and psychology, to support and enrich the knowledge that is learned; (c) harness compositionality and learning-to-learn to rapidly acquire and generalize knowledge to new tasks and situations.
Human
Ingredient
start-up software
(Infants):intuitive physics (primitive object concepts, Causality)
intuitive psychology (goals and beliefs)
Learning
model building : explaining observed data through the construction of causal models of the world
compositionality : capable of learning these richly structured models
from very limited amounts of experience
learning to learn : make this type of rapid model learning possible
Action 1. rich models our minds build are put into action, in real time. 2.
learn to do inference
: model-free methods can accelerate slow model-based inferences in perception and cognitionIntegration of model-based and model-free methods in reinforcement learning.
Explore Challenges
Recognizing new characters and objects
Learning to play the game Frostbite.
The ingredients outlined in this article will prove useful for working towards this goal: seeing objects and agents rather than features, building causal models and not just recognizing patterns, recombining representations without needing to retrain, and learning-to-learn rather than starting from scratch.
Last updated