Design principles for lifelong learning AI accelerators
Lifelong learning—an agent's ability to learn throughout its lifetime—is a hallmark of
biological learning systems and a central challenge for artificial intelligence (AI). The …
biological learning systems and a central challenge for artificial intelligence (AI). The …
Contributions by metaplasticity to solving the catastrophic forgetting problem
Catastrophic forgetting (CF) refers to the sudden and severe loss of prior information in
learning systems when acquiring new information. CF has been an Achilles heel of standard …
learning systems when acquiring new information. CF has been an Achilles heel of standard …
Memristor neural network circuit based on operant conditioning with immediacy and satiety
J Sun, Y Wang, P Liu, Y Wang - IEEE Transactions on …, 2022 - ieeexplore.ieee.org
Most of the operant conditioning only consider the basic theory, but the influencing factors
such as immediacy and satiety are ignored. In this paper, a memristor neural network circuit …
such as immediacy and satiety are ignored. In this paper, a memristor neural network circuit …
Continual Learning and Catastrophic Forgetting
This book chapter delves into the dynamics of continual learning, which is the process of
incrementally learning from a non-stationary stream of data. Although continual learning is a …
incrementally learning from a non-stationary stream of data. Although continual learning is a …
Scolar: A spiking digital accelerator with dual fixed point for continual learning
Spiking neural network models when deployed in dynamic environments, catastrophically
forget previously learned tasks. In this paper, we propose a reconfigurable spiking digital …
forget previously learned tasks. In this paper, we propose a reconfigurable spiking digital …
Metaplasticnet: Architecture with probabilistic metaplastic synapses for continual learning
Metaplasticity, the activity-dependent modification of synaptic plasticity, is an important
technique for mitigating catastrophic forgetting in neural networks. Often, continual learning …
technique for mitigating catastrophic forgetting in neural networks. Often, continual learning …