Learning Multi-Sensory Integration with Self- Organization and Statistics Johannes Bauer, Stefan Wermter http://www.informatik.uni-hamburg.de/WTM/
The Superior Colliculus Johannes Bauer Multi-Sensory Integration through Self-Organization and Statistics 2
The Superior Colliculus inverse effectiveness [1] spatial principle [1] sub-additive coincident neural response close additive disparate super-additive uni-sensory baseline optimal integration [2] Johannes Bauer Multi-Sensory Integration through Self-Organization and Statistics 3
Whatβs Interesting About That? ? Place in SC Meaning of Input Johannes Bauer Multi-Sensory Integration through Self-Organization and Statistics 4
The Algorithm [3] π: input stimulus π π : activity of π π π π : preferred value of π π Johannes Bauer Multi-Sensory Integration through Self-Organization and Statistics 5
The Algorithm [3] π: input stimulus π π : activity of π π π π : preferred value of π π βΌ π π 1 , π 2 , β¦ , π π π = π π ) π π = π π π 1 , π 2 , β¦ , π π π(π = π π ) π π 1 , π 2 , β¦ , π π Johannes Bauer Multi-Sensory Integration through Self-Organization and Statistics 6
The Algorithm [3] π: input stimulus π π : activity of π π π π : preferred value of π π βΌ π π 1 , π 2 , β¦ , π π π = π π ) π π = π π π 1 , π 2 , β¦ , π π π(π = π π ) π π 1 , π 2 , β¦ , π π Johannes Bauer Multi-Sensory Integration through Self-Organization and Statistics 6
The Algorithm [3] π: input stimulus π π : activity of π π π π : preferred value of π π βΌ π(π π β£ π = π π ) π π π = π π π 1 , π 2 , β¦ , π π π(π = π π ) π(π π ) π Noise independent. Johannes Bauer Multi-Sensory Integration through Self-Organization and Statistics 6
The Algorithm [3] π: input stimulus π π : activity of π π π π : preferred value of π π βΌ π(π π β£ π = π π ) π π π = π π π 1 , π 2 , β¦ , π π π(π π ) π Noise independent. π unif. dist. Johannes Bauer Multi-Sensory Integration through Self-Organization and Statistics 6
The Algorithm [3] π: input stimulus π π : activity of π π π π : preferred value of π π π π = π π π 1 , π 2 , β¦ , π π βΌ π(π π β£ π = π π ) π Noise independent. π unif. dist. Johannes Bauer Multi-Sensory Integration through Self-Organization and Statistics 6
The Algorithm [3] π π = π π π 1 , π 2 , β¦ , π π βΌ π(π π β£ π = π π ) π Johannes Bauer Multi-Sensory Integration through Self-Organization and Statistics 11
The Algorithm [3] can be adapted SOM-like ππ π,π [π π ] Assume π π π β£ π = π π = βππ π,π , then π π = π π β£ π 1 , π 2 , β¦ , π π βΌ ππ π,π π π β ππ π,π π Johannes Bauer Multi-Sensory Integration through Self-Organization and Statistics 12
Comparison to regular SOM Johannes Bauer Multi-Sensory Integration through Self-Organization and Statistics 13
Comparison to regular SOM Johannes Bauer Multi-Sensory Integration through Self-Organization and Statistics 13
The Network in Action βvisualβ input βauditoryβ input population-coded PDF Johannes Bauer Multi-Sensory Integration through Self-Organization and Statistics 15
The Network in Action βvisualβ input βauditoryβ input population-coded PDF Johannes Bauer Multi-Sensory Integration through Self-Organization and Statistics 16
Performance The network integrates multi-sensory information. errors given βvisualβ input errors given βauditoryβ input errors given multi-sensory input The network replicates biological phenomena. Spatial Principle Inverse Effectiveness Johannes Bauer Multi-Sensory Integration through Self-Organization and Statistics 17
Conclusion presented a novel self-organizing ANN algorithm which ο§ learns to combine information near-optimally ο§ shows spatial principle and MLE-like behavior ο§ shows benefit of multisensory integration ο§ learns to compute a PDF for latent variables ο§ is unsupervised ο§ has few inbuilt assumptions Johannes Bauer Multi-Sensory Integration through Self-Organization and Statistics 18
The End References: [1]: Stanford, T. R., Quessy, S., Stein, B. E., Jul. 2005. Evaluating the operations underlying multisensory integration in the cat superior colliculus. The Journal of Neuroscience 25 (28), 6499 β 6508. [2]: Alais, D., Burr, D., Feb. 2004. The ventriloquist effect results from Near-Optimal bimodal integration . Current Biology 14 (3), 257 β 262. [3]: Bauer, J. and Wermter, S., Sept. 2013. Self-organized neural learning of statistical inference from high-dimensional data. In: Proceedings of the International Joint Conference on Artificial Intelligence 2013 . Johannes Bauer Multi-Sensory Integration through Self-Organization and Statistics 14
Performance β Behavioral * *simulation parameters differ from rest of talk. auditory π π = 4.594 β 10 β4 π π = 1.680 β 10 β1 visual π π€ = 1.061 β 10 β4 π π€ = 8.272 β 10 β1 optimal 1 β 4.185 β 10 β5 π π,πππ’ = 1 2 + 1 2 ππ€ ππ multi-sensory π π,πππ’ β 1.876 β 10 β1 π π = 8.153 β 10 β5 π π€,πππ’ β 8.124 β 10 β1 Johannes Bauer Multi-Sensory Integration through Self-Organization and Statistics 20
Recommend
More recommend