Meta-learning in humans, monkeys, and robots
Abstract
The brain does not simply learn from error, but appears to control error-sensitivity, that is, in some cases a given prediction error results in robust learning, whereas in other cases the same error produces little or no learning. How does the brain control how much it is willing to learn from error? Understanding control of error-sensitivity is important both from a biological perspective, and from a machine learning perspective. From a biological perspective, control of error-sensitivity may provide insights into two critical puzzles: savings and meta-learning. Savings refers to the observation that training in task (A), followed by washout, produces accelerated re-learning of (A). Meta-learning refers to the observation that training in task (A), followed by washout, produces accelerated learning of task (- A). From a machine learning perspective, control of error-sensitivity dominates rates of convergence of internal models, learning of trajectories, and reinforcement-dependent procedures. It also dictates whether learning of one task can benefit the machine’s ability to transfer learning to a related task. Here, we have formed a team to tackle the problem of learning from a new direction: the ability to control error-sensitivity through acquisition of a memory of errors. Our team has the expertise to explore this question in three domains: human psychophysics, monkey neurophysiology, and robotics.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Aug 12, 2016
- Source ID
- N000141512312
Entities
People
- Reza Shadmher
Organizations
- Johns Hopkins University
- Office of Naval Research
- United States Navy