Joint program aims to enable autonomous systems to learn for life
NewsJuly 23, 2018
MALIBU, Calif. HRL Laboratories and the Defense Advanced Research Project Agency (DARPA) are teaming to develop what could be a breakthrough in machine-learning architecture for autonomous systems.
With funding from the Lifelong Learning Machines (L2M) program -- led by Dr. Hava Siegelmann from DARPA -- the proposed system will continually improve its performance and update its knowledge based on experience, performing these tasks without human supervision.
If successful, the system -- dubbed Super Turing Evolving Lifelong Learning ARchitecture (STELLAR) project -- will be a significant improvement over current technology, enabling autonomous systems to rapidly adapt to unforeseen situations, remember and learn from each experience, and consolidate new tasks with previously performed ones. Currently used machine-learning systems forget old tasks when learning new ones and are incapable of learning online and responding to new situations that were not presented during offline training.
“We intend to mimic the learning consolidation process of the brain using agents augmented with a memory system that is inspired by the interactions between the neocortex and hippocampus,” said Dr. Praveen Pilly, HRL’s principal investigator on the STELLAR project.
Dr. Soheil Kolouri, STELLAR’s co-principal investigator, said, "Current machine learning algorithms, including deep neural networks, are uniformly plastic and do not include mechanisms for preserving previously learned knowledge. Inspired by the neuromodulatory systems in the adult human brain, we’ll incorporate mechanisms of structural and functional plasticity that are capable of continual learning."
The HRL team also has members at Stanford University, University of California Irvine, University of Texas Austin, Loughborough University (U.K.), IT University of Copenhagen, and The French Institute for Research in Computer Science and Automation.
HRL Laboratories is a corporate research-and-development laboratory owned by The Boeing Company and General Motors.
