Timezone: »

Neural Network Poisson Models for Behavioural and Neural Spike Train Data
Moein Khajehnejad · Forough Habibollahi · Richard Nock · Ehsan Arabzadeh · Peter Dayan · Amir Dezfouli

@ None #None

One of the most important and challenging application areas for complex machine learning methods is to predict, characterize and model rich, multi-dimensional, neural data. Recent advances in neural recording techniques have made it possible to monitor the activities of a large number of neurons across different brain regions as animals perform behavioural tasks. This poses the critical challenge of establishing links between neural activity at a microscopic scale, which might for instance represent sensory input, and at a macroscopic scale, which then generates behaviour. Predominant modeling methods apply rather disjoint techniques to these scales; by contrast, we suggest an end-to-end model which exploits recent developments of flexible, but tractable, neural network point-process models to characterize dependencies between stimuli, actions, and neural data. We apply this model to a public dataset collected using Neuropixel probes in mice performing a visually-guided behavioural task as well as a synthetic dataset produced from a hierarchical network model with reciprocally connected sensory and integration circuits intended to characterize animal behaviour in a fixed-duration motion dis- crimination task. We show that our model outperforms previous approaches and contributes novel insights into the relationships between neural activities and behaviour.

Author Information

Moein Khajehnejad (Monash University)
Forough Habibollahi (University of Melbourne)

I am a Ph.D. candidate at the Department of Biomedical Engineering, Melbourne Brain Centre, the University of Melbourne. I work under the supervision of Prof. Anthony Burkitt (NeuroEngineering Laboratory) and Dr. Chris French (Neural Dynamics Laboratory). My primary research interests include Criticality, Cognitive Neuroscience, Brain-Computer Interfaces, Statistical Machine Learning, and Complex Networks Analysis. My Ph.D. research involves calcium imaging recordings from freely behaving mice using Miniscopes as well as recordings from cultured neuronal networks integrated with traditional silicon computing in a simulated game environment of ’pong’. I am performing computational analysis to understand the underlying mechanisms of information processing and cognition in networks of neurons that are involved in a cognitive or behavioural task. With a focus on population doctrine during cognition, I am studying the emergence of "critical" dynamics in the hippocampal/in-vitro neuronal networks during cognitively engaging tasks. I am also a part-time data scientist at CorticalLabs which is a startup company based in Melbourne where we work with live biological neuronal cultures integrated with traditional silicon computing. Prior to initiating my PhD, I commenced a Master’s Degree in Bioengineering at École polytechnique fédérale de Lausanne (EPFL), Lausanne, Switzerland (2018). During this period, I was a member of the Gräff Lab at Brain Mind Institute. I completed my B.Sc. degree in Electrical Engineering from Sharif University of Technology (SUT), Tehran, Iran (2018). As an undergraduate, I was a visiting scholar at the Singapore University of Technology and Design (SUTD-MIT). I worked remotely with the Monash Clinical and Imaging Neuroscience Laboratory (Melbourne, Australia) to develop an open-source MATLAB toolbox for external control of transcranial magnetic stimulation devices (MAGIC).

Richard Nock (Google Research)
Ehsan Arabzadeh (ANU)
Peter Dayan (MPI for Biological Cybernetics)
Amir Dezfouli (CSIRO's Data61)

Related Events (a corresponding poster, oral, or spotlight)

More from the Same Authors