Time Organized Maps Learning cortical topography from spatiotemporal

  • Slides: 20
Download presentation
Time Organized Maps – Learning cortical topography from spatiotemporal stimuli “Learning cortical topography from

Time Organized Maps – Learning cortical topography from spatiotemporal stimuli “Learning cortical topography from spatiotemporal stimuli”, J. Wiemer, F. Spengler, F. Joublin, P. Stagge, S. Wacquant, Biological Cybernetics, 2000 “The Time-Organized Map Algorithm: Extending the Self-Organizing Map to Spatiotemporal Signals”, Jan C. Wiemer, Neural Computation, 2003 Presented by: Mojtaba Solgi

Outline 1. The purpose and biological motivation 2. The Model: TOM Algorithm Wave propagation

Outline 1. The purpose and biological motivation 2. The Model: TOM Algorithm Wave propagation Learning • • Experiments and Results 3. Gaussian stimuli Generic artificial stimuli Semi-natural stimuli • • • 4. Discussion 5. z

Neurobiological experiments, Spengler et al. , 1996, 1999

Neurobiological experiments, Spengler et al. , 1996, 1999

Terminology Integration Fusion of different stimuli into one representation Segregation: Process of Increasing representational

Terminology Integration Fusion of different stimuli into one representation Segregation: Process of Increasing representational distance of different stimuli z

2 D Network Architecture Activation positional shift

2 D Network Architecture Activation positional shift

One-dimensional model

One-dimensional model

Wave propagation

Wave propagation

Integration and Segregation

Integration and Segregation

Algorithm 1. 2. Compute neurons activations and the position of the top winner neuron

Algorithm 1. 2. Compute neurons activations and the position of the top winner neuron Compute the neural position of the propagated wave from the last time step activation

Algorithm – Cont. 3. Shift the position of the top winner neuron due to

Algorithm – Cont. 3. Shift the position of the top winner neuron due to interaction with propagated wave

Algorithm – Cont. 4. 5. Again shift the position of the winner neuron this

Algorithm – Cont. 4. 5. Again shift the position of the winner neuron this time due to noise Update the winner neurons weights SOM Hebbian

Experiments with Gaussian stimuli & 2 D neural layer 1. Simulation of ‘ontogenesis’ (Development)

Experiments with Gaussian stimuli & 2 D neural layer 1. Simulation of ‘ontogenesis’ (Development)

Experiments with Gaussian stimuli & 2 D neural layer 2. Simulation of post-ontogenetic plasticity

Experiments with Gaussian stimuli & 2 D neural layer 2. Simulation of post-ontogenetic plasticity

One-dimensional model

One-dimensional model

Experiments with generic artificial stimuli & 1 D neural layer The input

Experiments with generic artificial stimuli & 1 D neural layer The input

Experiments with semi-natural stimuli & 1 D neural layer

Experiments with semi-natural stimuli & 1 D neural layer

Experiments with semi-natural stimuli & 1 D neural layer

Experiments with semi-natural stimuli & 1 D neural layer

Discussion Importance of temporal stimulus for development of cortical topography Continuous mapping of related

Discussion Importance of temporal stimulus for development of cortical topography Continuous mapping of related stimuli Inter-Stimulus-Interval-Dependant representations Hardly scalable No recognition performance on real-world problems Tested only on artificial input

Summary Utilizing temporal information in developing cortical topography Wave-like spread of cortical activity Experiments

Summary Utilizing temporal information in developing cortical topography Wave-like spread of cortical activity Experiments and results show compatibility of the model with neurobiological observations Biologically inspired and plausible, but no engineering performance

Thank you! Any thoughts/question?

Thank you! Any thoughts/question?