On-the-Fly Learning in a Perpetual Learning Machine

2015/09/03 by Andrew J. R. Simpson, Simpson, Andrew J. R. · 1 voice · 1 citation
Computer Science · #68Txx #FOS: Computer and information sciences #Machine Learning (cs.LG) #cs.LG #msc:68Txx

paper · pdf · doi:10.48550/arxiv.1509.00913

arxiv published 2015/09/03 · arxiv created 2015/09/29 · arxiv updated 2015/09/29

Abstract

Despite the promise of brain-inspired machine learning, deep neural networks (DNN) have frustratingly failed to bridge the deceptively large gap between learning and memory. Here, we introduce a Perpetual Learning Machine; a new type of DNN that is capable of brain-like dynamic 'on the fly' learning because it exists in a self-supervised state of Perpetual Stochastic Gradient Descent. Thus, we provide the means to unify learning and memory within a machine learning framework. We also explore the elegant duality of abstraction and synthesis: the Yin and Yang of deep learning.

Cited by

• Use it or Lose it: Selective Memory and Forgetting in a Perpetual Learning Machine

Discussions

• Perpetual Learning Machine [hn, 1 points, 0 comments]

Related

• Use it or Lose it: Selective Memory and Forgetting in a Perpetual Learning Machine
• Towards continual task learning in artificial neural networks: current approaches and insights from neuroscience
• Cognitively Inspired Learning of Incremental Drifting Concepts
• Semi-parametric Memory Consolidation: Towards Brain-like Deep Continual Learning
• Continual Learning with Deep Learning Methods in an Application-Oriented Context