Meta-learning local synaptic plasticity for continual familiarity detection
Preprint
Abstract
Over the course of a lifetime, a continual stream of information is encoded and retrieved from memory. To explore the synaptic mechanisms that enable this ongoing process, we consider a continual familiarity detection task in which a subject must report whether an image has been previously encountered. We design a class of feedforward neural network models endowed with biologically plausible synaptic plasticity dynamics, the parameters of which are meta-learned to optimize familiarity detection over long delay intervals. After training, we find that anti-Hebbian plasticity leads to better performance than Hebbian and replicates experimental results from the inferotemporal cortex, including repetition suppression. Unlike previous models, this network both operates continuously without requiring any synaptic resets and generalizes to intervals it has not been trained on. We demonstrate this not only for uncorrelated random stimuli but also for images of real-world objects. Our work suggests a biologically plausible mechanism for continual learning, and demonstrates an effective application of machine learning for neuroscience discovery.Keywords
All Related Versions
- Published version: Neuron, 110 (3), 544.
This publication has 36 references indexed in Scilit:
- Visual long-term memory has a massive storage capacity for object detailsProceedings of the National Academy of Sciences of the United States of America, 2008
- Computational models can replicate the capacity of human recognition memoryNetwork: Computation in Neural Systems, 2008
- Cascade Models of Synaptically Stored MemoriesNeuron, 2005
- Modeling hippocampal and neocortical contributions to recognition memory: A complementary-learning-systems approach.Psychological Review, 2003
- Comparison of computational models of familiarity discrimination in the perirhinal cortexHippocampus, 2003
- Long Short-Term MemoryNeural Computation, 1997
- Inferior Temporal Mechanisms for Invariant Object RecognitionCerebral Cortex, 1994
- Learning representations by back-propagating errorsNature, 1986
- A memory which forgetsJournal of Physics A: General Physics, 1986
- An analysis of the strength-latency relationshipMemory & Cognition, 1985