computa on through dynamics
play

Computa(on through dynamics Using recurrent neural networks to - PowerPoint PPT Presentation

Feb 21, 2023 •503 likes •1.57k views

Computa(on through dynamics Using recurrent neural networks to unveil mechanism in neural circuits David Sussillo with Valerio Mante and Bill Newsome Table of contents Introduc(on Training recurrent neural networks(RNNs) Understanding how

  1. Fixed points make a line amractor choice color le; choice choice right

  2. Two line amractors for two contexts context sensory evidence color e c i o h c motion The line amractors are context dependent and never exist at the same (me.

  3. Fixed points make a line amractor choice color le; choice choice right

  4. A simulated perturba(on experiment choice color context le; choice right

  5. A simulated perturba(on experiment choice color context le; choice right

  6. A simulated perturba(on experiment choice color context le; choice right

  7. A simulated perturba(on experiment choice color context le; choice right

  8. A simulated perturba(on experiment choice color context le; choice right

  9. A simulated perturba(on experiment choice color context le; choice right

  10. A simulated perturba(on experiment choice color context le; choice right

  11. A simulated perturba(on experiment choice color context le; choice right

  12. A simulated perturba(on experiment choice color context le; choice right

  13. A simulated perturba(on experiment choice color context le; choice right

  14. So what causes this difference between integra(on of color and ignoring mo(on? choice color context le; choice right

  15. Projec(ons onto the line amractor

  17. These vectors we’ve talked about are context independent

  19. The dynamics are context dependent

  20. The dynamics are context dependent

  21. The dynamics are context dependent

  22. The dynamics are context dependent

  23. The dynamics are context dependent

  24. The dynamics are context dependent

  25. The dynamics are context dependent

  26. The dynamics are context dependent

  27. The dynamics are context dependent

  28. The dynamics are context dependent

  29. The dynamics are context dependent

  30. The dynamics are context dependent

  31. The dynamics are context dependent

  32. The dynamics are context dependent

  33. Flexible selec(on and integra(on

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

DistributedAlgorithms computa@onandmessagedelivery Asynchronous PART

DistributedAlgorithms computa@onandmessagedelivery Asynchronous PART

3/4/09 Distributed AsynchronousSystems Bertinoro, March 2009 Algorithms Part II 2 Synchronous lockstepsynchrony,rounds DistributedAlgorithms computa@onandmessagedelivery

242 views • 22 slides
Randomness and quantum computa/on Computability and the BSS

Randomness and quantum computa/on Computability and the BSS

Randomness and quantum computa/on Computability and the BSS model, Hiddensee, August 9 , 2016 Andr Nies University of Auckland

652 views • 37 slides
Introduc)onto Computa)onal LexicalSeman)cs BillMacCartney

Introduc)onto Computa)onal LexicalSeman)cs BillMacCartney

Introduc)onto Computa)onal LexicalSeman)cs BillMacCartney CS224U,Lecture2 StanfordUniversity 12January2012 [slidesadaptedfromDanJurafsky] Outline 1)

936 views • 75 slides
The Computa,onal Requirements of Centralized RAN Matthew C. Valenti,

The Computa,onal Requirements of Centralized RAN Matthew C. Valenti,

The Computa,onal Requirements of Centralized RAN Matthew C. Valenti, West Virginia University Joint work with Peter Rost, Nokia Networks Aleksandra Checko, MTI Radiocomp Main Question Addressed in this Talk In a

356 views • 32 slides
Computa@onal Frameworks for Seman@c Analysis and Wikifica@on

Computa@onal Frameworks for Seman@c Analysis and Wikifica@on

Computa@onal Frameworks for Seman@c Analysis and Wikifica@on Dan Roth Department of Computer Science University of Illinois at Urbana-Champaign With

757 views • 58 slides
Web Dynamics Part 1 - Introduction 1.1 Dimensions of dynamics in the Web 1.2 Application

Web Dynamics Part 1 - Introduction 1.1 Dimensions of dynamics in the Web 1.2 Application

Web Dynamics Part 1 - Introduction 1.1 Dimensions of dynamics in the Web 1.2 Application examples Summer Term 2010 Web Dynamics 1-1 Why Web Dynamics? From Wikipedia: In physics the term dynamics customarily refers to the time evolution of

440 views • 42 slides
Mul$-scale Computa$onal Explora$on of Two-Dimensional Materials

Mul$-scale Computa$onal Explora$on of Two-Dimensional Materials

Mul$-scale Computa$onal Explora$on of Two-Dimensional Materials in Nanofluidics and DNA Sequencing PI: Narayana R. Aluru Presenter: Yanbin Wu Department

561 views • 11 slides
WHY? 1 26/01/18 1. Can be used to simulate molecular motion in a viscous medium, without

WHY? 1 26/01/18 1. Can be used to simulate molecular motion in a viscous medium, without

26/01/18 ADVANCED TECHNIQUES (MC/MD) A (seemingly) random selection. Daan Frenkel Beyond Newtonian MD 1. Langevin dynamics 2. Brownian dynamics 3. Stokesian dynamics 4. Dissipative particle dynamics 5. Etc. etc. WHY? 1 26/01/18 1. Can be used

994 views • 48 slides
10-601 Machine Learning HMM applica)ons in computa)onal biology Central

10-601 Machine Learning HMM applica)ons in computa)onal biology Central

10-601 Machine Learning HMM applica)ons in computa)onal biology Central dogma CCTGAGCCAACTATTGATGAA DNA transcription mRNA CCUGAGCCAACUAUUGAUGAA translation Protein PEPTIDE 2 Biological data

869 views • 45 slides
Feb 18, 2015 Supported by EU FP7-IRSES projects Computa(onal Intelligence

Feb 18, 2015 Supported by EU FP7-IRSES projects Computa(onal Intelligence

A Brief Introduction Feb 18, 2015 Supported by EU FP7-IRSES projects Computa(onal Intelligence Lab (CIL) EYE2E (269118), LIVCODE (295151) University of Lincoln and HAZCEPT (318907). United

659 views • 17 slides
Computa(onal and Policy Tools for Reproducible Research Roger

Computa(onal and Policy Tools for Reproducible Research Roger

Computa(onal and Policy Tools for Reproducible Research Roger D. Peng, PhD Department of Biosta/s/cs Johns Hopkins Bloomberg School of Public Health July

608 views • 46 slides
WHY? 1 17/01/17 1. Can be used to simulate molecular motion in a viscous medium, without

WHY? 1 17/01/17 1. Can be used to simulate molecular motion in a viscous medium, without

17/01/17 ADVANCED TECHNIQUES (MC/MD) A (seemingly) random selection. Daan Frenkel U. Cambridge 13/01/2017 But first: beyond Newtonian MD 1. Langevin dynamics 2. Brownian dynamics 3. Stokesian dynamics 4. Dissipative particle dynamics 5. Etc.

696 views • 47 slides
Mo#f discovery Morgane Thomas-Chollier Computa)onal systems

Mo#f discovery Morgane Thomas-Chollier Computa)onal systems

Mo#f discovery Morgane Thomas-Chollier Computa)onal systems biology - IBENS mthomas@biologie.ens.fr M2 Computa6onal analysis of cis-regulatory sequences

527 views • 41 slides
WHY? 1 <latexit

WHY? 1 <latexit

15/01/19 ADVANCED TECHNIQUES (MC/MD) A (seemingly) random selection. Daan Frenkel Beyond Newtonian MD 1. Langevin dynamics 2. Brownian dynamics 3. Stokesian dynamics 4. Dissipative particle dynamics 5. Etc. etc. WHY? 1 <latexit

818 views • 53 slides
Web Dynamics Part 1 Introduction 1.1 Dimensions of dynamics in the Web 1.2 Application examples

Web Dynamics Part 1 Introduction 1.1 Dimensions of dynamics in the Web 1.2 Application examples

Web Dynamics Part 1 Introduction 1.1 Dimensions of dynamics in the Web 1.2 Application examples Summer Term 2009 Web Dynamics 1 1 Why Web Dynamics? From Wikipedia: In physics the term dynamics customarily refers to the time evolution of

608 views • 35 slides
Complemen(ng*Computa(on*with* Visualiza(on*in*Genomics* March*11,*2010*

Complemen(ng*Computa(on*with* Visualiza(on*in*Genomics* March*11,*2010*

British Columbia Cancer Agency Genome Sciences Centre Vancouver . British Columbia . Canada Complemen(ng*Computa(on*with* Visualiza(on*in*Genomics* March*11,*2010* EBI*Interfaces*Interest*Forum* Cydney*Nielsen* Discovery*path*

483 views • 45 slides
Learning to optimize multigrid PDE solvers DANIEL GREENFELD, WEIZMANN INSTITUTE OF SCIENCE JOINT

Learning to optimize multigrid PDE solvers DANIEL GREENFELD, WEIZMANN INSTITUTE OF SCIENCE JOINT

Learning to optimize multigrid PDE solvers DANIEL GREENFELD, WEIZMANN INSTITUTE OF SCIENCE JOINT WORK W. MEIRAV GALUN, RON KIMMEL, IRAD YAVNEH AND RONEN BASRI Solving PDEs is useful Predicting weather systems Aircraft and auto design

453 views • 9 slides
Apixaban versus Heparin/Vitamin K Antagonist in Anticoagulation-nave Patients with Atrial

Apixaban versus Heparin/Vitamin K Antagonist in Anticoagulation-nave Patients with Atrial

Apixaban versus Heparin/Vitamin K Antagonist in Anticoagulation-nave Patients with Atrial Fibrillation Scheduled for Cardioversion: The EMANATE Trial Michael D. Ezekowitz, Professor, Sidney Kimmel Medical College at Thomas Jefferson

354 views • 16 slides
Simultaneous Multiparty Communication Protocols for Composed Functions Yassine Hamoudi IRIF ,

Simultaneous Multiparty Communication Protocols for Composed Functions Yassine Hamoudi IRIF ,

Simultaneous Multiparty Communication Protocols for Composed Functions Yassine Hamoudi IRIF , Universit Paris Diderot, CNRS MFCS 2018 Number-On-Forehead model [Chandra, Furst, Lipton83] 2 F : X 1 X k {0,1} Player 1 Player

772 views • 33 slides
Theoretical insights on pinning Ivan Sadovskyy (Argonne, UChicago)

Theoretical insights on pinning Ivan Sadovskyy (Argonne, UChicago)

Theoretical insights on pinning Ivan Sadovskyy (Argonne, UChicago) Andreas Glatz (Argonne, NIU) Alexei Koshelev (Argonne) Gregory Kimmel (Northwestern) TTC Topical Workshop,

444 views • 18 slides
Back to School Night AE Wright Middle School Thank You Board Members! LVUSD Cabinet Guests

Back to School Night AE Wright Middle School Thank You Board Members! LVUSD Cabinet Guests

Back to School Night AE Wright Middle School Thank You Board Members! LVUSD Cabinet Guests Dr. Dan Stepenosky, Superintendent Dr. Rose Dunn, Assistant Superintendent, Personnel Dr. Clara Finneran, Assistant Superintendent, Education

726 views • 13 slides
Ap Applications of the Quantum st st-Co Connectivity y Algori rithm June 3, 2019 University

Ap Applications of the Quantum st st-Co Connectivity y Algori rithm June 3, 2019 University

Ap Applications of the Quantum st st-Co Connectivity y Algori rithm June 3, 2019 University of Maryland Kai DeLorenzo 1 , Shelby Kimmel 1 , and R. Teal Witter 1 1 Middlebury College La Layers of of A Abstract ction on Our Algorithms

251 views • 23 slides
Convex relaxations for weakly supervised information extraction Edouard Grave Columbia

Convex relaxations for weakly supervised information extraction Edouard Grave Columbia

Convex relaxations for weakly supervised information extraction Edouard Grave Columbia University edouard.grave@gmail.com Information Extraction Extract structured information from unstructured documents. Information Extraction Extract

1.03k views • 81 slides
Cosmological Hydrodynamic Simulations Andrew Wetzel SUMMARY OF THIS TALK Cosmological

Cosmological Hydrodynamic Simulations Andrew Wetzel SUMMARY OF THIS TALK Cosmological

Cosmological Hydrodynamic Simulations Andrew Wetzel SUMMARY OF THIS TALK Cosmological hydrodynamic simulations are the most powerful theoretical tools to study stellar halos you just have to solve galaxy formation first. Andrew Wetzel

717 views • 22 slides

More recommend