Markov Chain Hebbian Learning Algorithm With Ternary Synaptic Units

Abstract

In spite of remarkable progress in machine learning techniques, the state-of-the-art machine learning algorithms often keep machines from real-time learning (online learning) due, in part, to computational complexity in parameter optimization. As an alternative, a learning algorithm to train a memory in real time is proposed, named the Markov chain Hebbian learning algorithm. The algorithm pursues efficient use in memory during training in that: 1) the weight matrix has ternary elements (-1, 0, 1) and 2) each update follows a Markov chain-the upcoming update does not need past weight values. The algorithm was verified by two proof-of-concept tasks: image (MNIST and CIFAR-10 datasets) recognition and multiplication table memorization. Particularly, the latter bases multiplication arithmetic on memory, which may be analogous to humans' mental arithmetic. The memory-based multiplication arithmetic feasibly offers the basis of factorization, supporting novel insight into memory-based arithmetic.

Keywords

• Greedy edge-wise training
• Hebbian learning
• Markov chain
• mental arithmetic
• prime factorization
• supervised learning