a unified theory for the origin of grid cells through the
play

A unified theory for the origin of grid cells through the lens of - PowerPoint PPT Presentation

A unified theory for the origin of grid cells through the lens of pattern formation Ben Sorscher*, Gabriel C. Mel*, Surya Ganguli, Sam Ocko Grid cells Krupic et al. (2012), Derdikman (2014) Trained neural networks learn grid patterns Banino et


  1. A unified theory for the origin of grid cells through the lens of pattern formation Ben Sorscher*, Gabriel C. Mel*, Surya Ganguli, Sam Ocko

  2. Grid cells Krupic et al. (2012), Derdikman (2014)

  3. Trained neural networks learn grid patterns Banino et al. (2018) Cueva & Wei (2018) unconstrained Dordek et al. (2016) Stachenfeld et al. (2014) nonnegative

  4. Trained neural networks learn grid patterns

  5. 1. Why are the optimal maps grids? 2. What determines the optimal grid type - square, amorphous, or hexagonal?

  6. Gradient descent as a pattern forming dynamics

  7. Gradient descent as a pattern forming dynamics

  8. Gradient descent as a pattern forming dynamics 1. Why are the optimal maps grids?

  9. Gradient descent as a pattern forming dynamics 1. Why are the optimal maps grids? Translation invariance => Fourier modes

  10. 1. Why are the optimal maps grids? Translation invariance => Fourier modes 2. What determines the optimal grid type - square, amorphous, or hexagonal?

  11. 1. Why are the optimal maps grids? Translation invariance => Fourier modes 2. What determines the optimal grid type - square, amorphous, or hexagonal?

  12. Nonnegativity yields hexagonal grids

  13. Nonnegativity yields hexagonal grids

  14. Nonnegativity yields hexagonal grids Taylor expand constraint

  15. Nonnegativity yields hexagonal grids Taylor expand constraint “Three-body interaction” between stripes 60 o apart

  16. Nonnegativity yields hexagonal grids Taylor expand constraint “Three-body interaction” between stripes 60 o apart

  17. Unifying mechanistic and normative models Grid cell model RNNs Normative encoding models Skaggs et al. (1995) Banino et al. (2018) Zhang (1996) Cueva & Wei (2018) Fuhs and Touretzky (2006) Dordek et al. (2016) Burak and Fiete (2009) Stachenfeld et al. (2014)

  18. Unifying mechanistic and normative models Grid cell model RNNs Normative encoding models Gradient Activity descent dynamics Skaggs et al. (1995) Banino et al. (2018) Zhang (1996) Cueva & Wei (2018) Fuhs and Touretzky (2006) Dordek et al. (2016) Burak and Fiete (2009) Stachenfeld et al. (2014) Pattern forming dynamics

  19. Unifying mechanistic and normative models Grid cell model RNNs Normative encoding models Gradient Activity = descent dynamics Skaggs et al. (1995) Banino et al. (2018) Zhang (1996) Cueva & Wei (2018) Fuhs and Touretzky (2006) Dordek et al. (2016) Burak and Fiete (2009) Stachenfeld et al. (2014) Pattern forming dynamics

Download Presentation
Download Policy: The content available on the website is offered to you 'AS IS' for your personal information and use only. It cannot be commercialized, licensed, or distributed on other websites without prior consent from the author. To download a presentation, simply click this link. If you encounter any difficulties during the download process, it's possible that the publisher has removed the file from their server.

Recommend


More recommend