Burst-Dependent Plasticity and Dendritic Amplification Support Target-Based Learning and Hierarchical Imitation Learning

Cristiano Capone, Cosimo Lupo, Paolo Muratore, Pier Stanislao Paolucci
Proceedings of the 39th International Conference on Machine Learning, PMLR 162:2625-2637, 2022.

Abstract

The brain can learn to solve a wide range of tasks with high temporal and energetic efficiency. However, most biological models are composed of simple single-compartment neurons and cannot achieve the state-of-the-art performances of artificial intelligence. We propose a multi-compartment model of pyramidal neuron, in which bursts and dendritic input segregation give the possibility to plausibly support a biological target-based learning. In target-based learning, the internal solution of a problem (a spatio-temporal pattern of bursts in our case) is suggested to the network, bypassing the problems of error backpropagation and credit assignment. Finally, we show that this neuronal architecture naturally supports the orchestration of “hierarchical imitation learning”, enabling the decomposition of challenging long-horizon decision-making tasks into simpler subtasks.

Cite this Paper


BibTeX
@InProceedings{pmlr-v162-capone22b, title = {Burst-Dependent Plasticity and Dendritic Amplification Support Target-Based Learning and Hierarchical Imitation Learning}, author = {Capone, Cristiano and Lupo, Cosimo and Muratore, Paolo and Paolucci, Pier Stanislao}, booktitle = {Proceedings of the 39th International Conference on Machine Learning}, pages = {2625--2637}, year = {2022}, editor = {Chaudhuri, Kamalika and Jegelka, Stefanie and Song, Le and Szepesvari, Csaba and Niu, Gang and Sabato, Sivan}, volume = {162}, series = {Proceedings of Machine Learning Research}, month = {17--23 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v162/capone22b/capone22b.pdf}, url = {https://proceedings.mlr.press/v162/capone22b.html}, abstract = {The brain can learn to solve a wide range of tasks with high temporal and energetic efficiency. However, most biological models are composed of simple single-compartment neurons and cannot achieve the state-of-the-art performances of artificial intelligence. We propose a multi-compartment model of pyramidal neuron, in which bursts and dendritic input segregation give the possibility to plausibly support a biological target-based learning. In target-based learning, the internal solution of a problem (a spatio-temporal pattern of bursts in our case) is suggested to the network, bypassing the problems of error backpropagation and credit assignment. Finally, we show that this neuronal architecture naturally supports the orchestration of “hierarchical imitation learning”, enabling the decomposition of challenging long-horizon decision-making tasks into simpler subtasks.} }
Endnote
%0 Conference Paper %T Burst-Dependent Plasticity and Dendritic Amplification Support Target-Based Learning and Hierarchical Imitation Learning %A Cristiano Capone %A Cosimo Lupo %A Paolo Muratore %A Pier Stanislao Paolucci %B Proceedings of the 39th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2022 %E Kamalika Chaudhuri %E Stefanie Jegelka %E Le Song %E Csaba Szepesvari %E Gang Niu %E Sivan Sabato %F pmlr-v162-capone22b %I PMLR %P 2625--2637 %U https://proceedings.mlr.press/v162/capone22b.html %V 162 %X The brain can learn to solve a wide range of tasks with high temporal and energetic efficiency. However, most biological models are composed of simple single-compartment neurons and cannot achieve the state-of-the-art performances of artificial intelligence. We propose a multi-compartment model of pyramidal neuron, in which bursts and dendritic input segregation give the possibility to plausibly support a biological target-based learning. In target-based learning, the internal solution of a problem (a spatio-temporal pattern of bursts in our case) is suggested to the network, bypassing the problems of error backpropagation and credit assignment. Finally, we show that this neuronal architecture naturally supports the orchestration of “hierarchical imitation learning”, enabling the decomposition of challenging long-horizon decision-making tasks into simpler subtasks.
APA
Capone, C., Lupo, C., Muratore, P. & Paolucci, P.S.. (2022). Burst-Dependent Plasticity and Dendritic Amplification Support Target-Based Learning and Hierarchical Imitation Learning. Proceedings of the 39th International Conference on Machine Learning, in Proceedings of Machine Learning Research 162:2625-2637 Available from https://proceedings.mlr.press/v162/capone22b.html.

Related Material