T1. 立命館大学 VLSIセンター(協賛) Vassilis Cutsuridis 准教授 講演会

Asso.Prof. Vassilis Cutsuridis / University of Lincoln, UK

https://staff.lincoln.ac.uk/dc36f34d-567d-4205-8ce2-42f28166ac0a

立命館大学 びわこ・くさつキャンパス ローム記念館5階大会議室
参加無料・事前申込み不要・対面のみ
7/9(火) 16:00～17:00
VLSIセンターの発表

Title : Neuromorphic associative memories

Abstract

Mathematical artificial neural network models known as associative memories first developed over 60 years ago worked by storing patterns of information via changes in the strengths of connections between simple computing units that mimicked the operation of neurons in the brain in a very simple way. Old memories were recalled when a noisy or partial version of a previously stored patterns was presented to the network. However, these devices were not very flexible. They had to be told when to store a pattern and when they were supposed to recall a memory. The types of information they could store were quite limited.