Jacob Huth (PhD student)

Vision Institute
Aging in Vision and Action Lab
CNRS – INSERM – University Pierre&Marie Curie
17, rue Moreau F-75012 Paris, France
Phone: +33 (0)1 53 46 25 61

email:

My interests lie in the intersection of modeling and computation. Models of neural and cognitive processes are very important to verify hypotheses about real wold systems. Recurrent and dynamic processes like neural networks (even simulated by very simple spiking models) can give rise to self organization principles and adaptive behaviour, which can enable a sufficiently large assembly of simple units to do complex computations, classifications and predictions. Our understanding of how neural computation in actual biological systems is used and can be used is dependent on our understanding of computation on a theoretical level. To find theories that are  both biologically plausible and explain sufficiently how and why the process adapts to a certain behaviour, a constant exchange is necessary.

In my current projects I am investigating the effect of aging on the human visual system by modeling a very simplified version of the early stages of visual processing and matching the behavior of the model to experimental data. The goal is to find the simplest explanatory variable that predicts a deterioration in cortical vision (correcting for optical and retinal differences). This variable will be then examined in further experiments.

Keywords: spiking neural models, visual processing, visual perception, visual aging.

Short CV:

Master of Science in Cognitive Science (University of Osnabrück) – Jan 2014
Bachelor of Science in Cognitive Science (University of Osnabrück) – Nov 2011

Miscellaneous:

What is Cognitive Science? – A film accompanying the Handbuch Kognitionswissenschaft by Simon Harst, Ngan Tram Ho Dac, Jacob Huth, Jonas Klein, Alexander Krüger, Alexander Niedrig, Johannes Merkel, Danja Porada, Laura Schmitz, Sebastian Timmer
https://www.youtube.com/watch?v=Tw3ims19L0g

 

CV (pdf): [ download ]

 

Publications

Order by Year - Category

Show All - Journals - Peer-reviewed conference proceedings - Others


2018

  1. Huth J, Masquelier T and Arleo A (2018) Convis: A Toolbox to Fit and Simulate Filter-Based Models of Early Visual Processing. Frontiers in Neuroinformatics, 12:9.
  2. Huth J (2018) Modelling Aging in the Visual System & the Convis Python Toolbox. Ph.D. Thesis, Sorbonne University, Paris.
  3. Huth J, Masquelier T and Arleo A (2018) What has Deep Learning ever done for us? Simulating the Visual System with convis. In 11th FENS Forum 2018.

2017

  1. Huth J, Masquelier T and Arleo A (2017) The convis toolbox: Population Simulation of the Visual System with Automatic Differentiation using Theano. In CNS.
  2. Huth J, Masquelier T and Arleo A (2017) The convis toolbox: Population Simulation of the Visual System with Automatic Differentiation using Theano. In Bernstein Conference.
  3. Huth J, Mayer E, Masquelier T and Arleo A (2017) Effects of Noise on Contrast Gain Control. In European Retina Meeting.
  4. Huth J, Masquelier T and Arleo A (2017) Convis: A Toolbox To Fit and Simulate Filter-based Models of Early Visual Processing. bioRxiv:169284.

2016

  1. Huth J (2016) A model of aging effects on the visual system. In NeWS Neuroscience Workshop Saclay.
  2. Huth J, Masquelier T, Allard R and Arleo A (2016) Modeling the Visual System with respect to Aging. In IK Interdisciplinary College.
  3. Huth J, Masquelier T, Allard R and Arleo A (2016) Using Equivalent Internal Noise as a Constraint on Vision Models. In Bernstein Conference.

2014

  1. Huth J (2014) Bootstrapping Information Criterion for Stochastic Models of Point Processes & The ni. Python Toolbox. Master's Thesis, University of Osnabrück.

2013

  1. Huth J (2013) Handbuch Kognitionswissenschaft. ISBN 3476023311, Metzler.

2011

  1. Huth J (2011) Synchrony Binding in Noise Excitable Coherence Resonance Networks. Master's Thesis, University of Osnabrück.