Computational principles of synaptic memory consolidation

Benna, Marcus K.; Fusi, Stefano

doi:10.1038/nn.4401

Cited by 202 publications

(357 citation statements)

References 45 publications

Supporting

Mentioning

304

Contrasting

Order By: Relevance

“…It does so by selectively decreasing the plasticity of weights and thus has certain parallels with neurobiological models of synaptic consolidation (15,16). We implement EWC as a soft, quadratic constraint whereby each weight is pulled back toward its old values by an amount proportional to its importance for performance on previously learned tasks.…”

Section: Discussionmentioning

confidence: 99%

“…Whereas this paper has primarily focused on building an algorithm inspired by neurobiological observations and theories (15,16), it is also instructive to consider whether the algorithm's successes can feed back into our understanding of the brain. In particular, we see considerable parallels between EWC and two computational theories of synaptic plasticity.…”

Section: Discussionmentioning

confidence: 99%

“…Cascade models of synaptic plasticity (15,16) construct dynamical models of synaptic states to understand the tradeoff between plasticity and memory retention. Cascade models have important differences from our approach.…”

Section: Discussionmentioning

confidence: 99%

“…Whereas prior work on cascade models (15,16) has tied the metaplastic state to patterns of potentiation and depression events-i.e., synaptic-level measures-our approach focuses on the computational principles that determine the degree to which each synapses might be consolidated. It may be possible to distinguish these models experimentally, because the plasticity of a synapse depends on the rate of potentiation events in the cascade model, but on task relevance in EWC.…”

Section: Discussionmentioning

confidence: 99%

“…This provides causal evidence that neural mechanisms supporting the protection of these strengthened synapses are critical to retention of task performance. These experimental findings-together with neurobiological models such as the cascade model (15,16)-suggest that continual learning in the neocortex relies on task-specific synaptic consolidation, whereby knowledge is durably encoded by rendering a proportion of synapses less plastic and therefore stable over long timescales.…”

Section: Overcoming Catastrophic Forgetting In Neural Networkmentioning

confidence: 93%

See 4 more Smart Citations

Overcoming catastrophic forgetting in neural networks

Kirkpatrick

Pascanu

Rabinowitz

et al. 2017

Proc. Natl. Acad. Sci. U.S.A.

4,444

3,870

View full text Add to dashboard Cite

The ability to learn tasks in a sequential fashion is crucial to the development of artificial intelligence. Until now neural networks have not been capable of this and it has been widely thought that catastrophic forgetting is an inevitable feature of connectionist models. We show that it is possible to overcome this limitation and train networks that can maintain expertise on tasks that they have not experienced for a long time. Our approach remembers old tasks by selectively slowing down learning on the weights important for those tasks. We demonstrate our approach is scalable and effective by solving a set of classification tasks based on a hand-written digit dataset and by learning several Atari 2600 games sequentially.synaptic consolidation | artificial intelligence | stability plasticity | continual learning | deep learning A chieving artificial general intelligence requires that agents are able to learn and remember many different tasks (1). This is particularly difficult in real-world settings: The sequence of tasks may not be explicitly labeled, tasks may switch unpredictably, and any individual task may not recur for long time intervals. Critically, therefore, intelligent agents must demonstrate a capacity for continual learning: that is, the ability to learn consecutive tasks without forgetting how to perform previously trained tasks.Continual learning poses particular challenges for artificial neural networks due to the tendency for knowledge of the previously learned task(s) (e.g., task A) to be abruptly lost as information relevant to the current task (e.g., task B) is incorporated. This phenomenon, termed catastrophic forgetting (2-6), occurs specifically when the network is trained sequentially on multiple tasks because the weights in the network that are important for task A are changed to meet the objectives of task B. Whereas recent advances in machine learning and in particular deep neural networks have resulted in impressive gains in performance across a variety of domains (e.g., refs. 7 and 8), little progress has been made in achieving continual learning. Current approaches have typically ensured that data from all tasks are simultaneously available during training. By interleaving data from multiple tasks during learning, forgetting does not occur because the weights of the network can be jointly optimized for performance on all tasks. In this regime-often referred to as the multitask learning paradigm-deep-learning techniques have been used to train single agents that can successfully play multiple Atari games (9, 10). If tasks are presented sequentially, multitask learning can be used only if the data are recorded by an episodic memory system and replayed to the network during training. This approach [often called system-level consolidation (4, 5)] is impractical for learning large numbers of tasks, as in our setting it would require the amount of memories being stored and replayed to be proportional to the number of tasks. The lack of algorithms to support continual learning thus rema...

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Section: Overcoming Catastrophic Forgetting In Neural Networkmentioning

confidence: 93%

See 3 more Smart Citations

Overcoming catastrophic forgetting in neural networks

Kirkpatrick

Pascanu

Rabinowitz

et al. 2017

Proc. Natl. Acad. Sci. U.S.A.

4,444

3,870

View full text Add to dashboard Cite

show abstract

Bridging Biological and Artificial Neural Networks with Emerging Neuromorphic Devices: Fundamentals, Progress, and Challenges

et al. 2019

View full text Add to dashboard Cite

As the research on artificial intelligence booms, there is broad interest in brain‐inspired computing using novel neuromorphic devices. The potential of various emerging materials and devices for neuromorphic computing has attracted extensive research efforts, leading to a large number of publications. Going forward, in order to better emulate the brain's functions, its relevant fundamentals, working mechanisms, and resultant behaviors need to be re‐visited, better understood, and connected to electronics. A systematic overview of biological and artificial neural systems is given, along with their related critical mechanisms. Recent progress in neuromorphic devices is reviewed and, more importantly, the existing challenges are highlighted to hopefully shed light on future research directions.

show abstract

A Marr's Three‐Level Analytical Framework for Neuromorphic Electronic Systems

Guo

Zou

et al. 2021

Advanced Intelligent Systems

View full text Add to dashboard Cite

Neuromorphic electronics, an emerging field that aims for building electronic mimics of the biological brain, holds promise for reshaping the frontiers of information technology and enabling a more intelligent and efficient computing paradigm. As their biological brain counterpart, the neuromorphic electronic systems are complex, having multiple levels of organization. Inspired by David Marr's famous three‐level analytical framework developed for neuroscience, the advances in neuromorphic electronic systems are selectively surveyed and given significance to these research endeavors as appropriate from the computational level, algorithmic level, or implementation level. Under this framework, the problem of how to build a neuromorphic electronic system is defined in a tractable way. In conclusion, the development of neuromorphic electronic systems confronts a similar challenge to the one neuroscience confronts, that is, the limited constructability of the low‐level knowledge (implementations and algorithms) to achieve high‐level brain‐like (human‐level) computational functions. An opportunity arises from the communication among different levels and their codesign. Neuroscience lab‐on‐neuromorphic chip platforms offer additional opportunity for mutual benefit between the two disciplines.

show abstract

Computational principles of synaptic memory consolidation

Cited by 202 publications

References 45 publications

Overcoming catastrophic forgetting in neural networks

Overcoming catastrophic forgetting in neural networks

Bridging Biological and Artificial Neural Networks with Emerging Neuromorphic Devices: Fundamentals, Progress, and Challenges

A Marr's Three‐Level Analytical Framework for Neuromorphic Electronic Systems

Contact Info

Product

Resources

About