Biology subjects

Isomura, T.

Publications and source records attributed to Isomura, T..

4 recordsLinked to original sources

bioRxiv2018

Multi-context blind source separation by error-gated Hebbian rule

Animals need to adjust their inferences according to the context they are in. This is required for the multi-context blind source separation (BSS) task, where an agent needs to infer hidden sources from their context-dependent mixtures. The agent is expected to invert this mixing process for all contexts. Here, we show that a neural network that implements the error-gated Hebbian rule (EGHR) with sufficiently redundant sensory inputs can successfully learn this task. After training, the network can perform the multi-context BSS without further updating synapses, by retaining memories of all experienced contexts. Finally, if there is a common feature shared across contexts, the EGHR can extract it and generalize the task to even inexperienced contexts. This demonstrates an attractive use of the EGHR for dimensionality reduction by extracting common sources across contexts. The results highlight the utility of the EGHR as a model for perceptual adaptation in animals.

neuroscience↗

bioRxiv2018

In vitro neural networks minimise variational free energy

In this work, we address the neuronal encoding problem from a Bayesian perspective. Specifically, we ask whether neuronal responses in an in vitro neuronal network are consistent with ideal Bayesian observer responses under the free energy principle. In brief, we stimulated an in vitro cortical cell culture with stimulus trains that had a known statistical structure. We then asked whether recorded neuronal responses were consistent with variational message passing (i.e., belief propagation) based upon free energy minimisation (i.e., evidence maximisation). Effectively, this required us to solve two problems: first, we had to formulate the Bayes-optimal encoding of the causes or sources of sensory stimulation, and then show that these idealised responses could account for observed electrophysiological responses. We describe a simulation of an optimal neural network (i.e., the ideal Bayesian neural code) and then consider the mapping from idealised in silico responses to recorded in vitro responses. Our objective was to find evidence for functional specialisation and segregation in the in vitro neural network that reproduced in silico learning via free energy minimisation. Finally, we combined the in vitro and in silico results to characterise learning in terms of trajectories in a variational information plane of accuracy and complexity.

neuroscience↗

bioRxiv2018

Social intelligence model with multiple internal models

To exhibit social intelligence, animals have to recognize who they are communicating with. One way to make this inference is to select among multiple internal generative models of each conspecific. This induces an interesting problem: when receiving sensory input generated by a particular conspecific, how does an animal know which internal model to update? We consider a theoretical and neurobiologically plausible solution that enables inference and learning under multiple generative models by integrating active inference and (online) Bayesian model selection. This scheme fits sensory inputs under each generative model. Model parameters are then updated in proportion to the probability it could have generated the current input (i.e., model evidence). We show that a synthetic bird who employs the proposed scheme successfully learns and distinguishes (real zebra finch) birdsongs generated by several different birds. These results highlight the utility of having multiple internal models to make inferences in complicated social environments.

neuroscience↗

bioRxiv2017

Common features in plastic changes rather than constructed structures in recurrent neural network prefrontal cortex models

We have flexible control over our cognition depending on the context or surrounding environments. The prefrontal cortex (PFC) controls this cognitive flexibility; however, the detailed underlying mechanisms remain unclear. Recent developments in machine learning techniques have allowed simple recurrent neural network PFC models to perform human- or animal-like behavioral tasks. These systems allow us to acquire parameters, which we could not in biological experiments, for performing the tasks. We compared four models, in which a flexible cognition task, called context-dependent integration task, was performed; subsequently, we searched for common features. In all the models, we observed that high plastic synapses were concentrated in the small neuronal population and the more concentrated neuronal units contributed further to the performance. However, there were no common properties in the constructed structures. These results suggest that plastic changes can be more general and important to accomplish cognitive tasks than features of the constructed structures.

neuroscience↗