Decomposing reverse-mode automatic differentiation (LAFI 2021)

Who

Roy Frostig, Matthew J. Johnson, Dougal Maclaurin, Adam Paszke, Alexey Radul

Track

LAFI 2021

Time Zone

The program is currently displayed in (GMT+01:00) Amsterdam, Berlin, Bern, Rome, Stockholm, Vienna.

Use conference time zone: (GMT+01:00) Amsterdam, Berlin, Bern, Rome, Stockholm, ViennaSelect other time zone

The GMT offsets shown reflect the offsets at the moment of the conference.

Time Band

By setting a time band, the program will dim events that are outside this time window. This is useful for (virtual) conferences with a continuous program (with repeated sessions).
The time band will also limit the events that are included in the personal iCalendar subscription service.

Display full programSpecify a time band

Save

When

Sun 17 Jan 2021 18:51 - 19:04 at LAFI - Session 2 Chair(s): Dougal Maclaurin

File attachments

abstract (lafi21-decomposing.pdf)	135KiB
slides (lafi21-slides.pdf)	268KiB

Roy Frostig

Google Research

Matthew J. Johnson

Google Brain

United States

Dougal Maclaurin

Google Research

Adam Paszke

Google Research

Poland

Alexey Radul

Google Research

Time Zone

The program is currently displayed in (GMT+01:00) Amsterdam, Berlin, Bern, Rome, Stockholm, Vienna.

Use conference time zone: (GMT+01:00) Amsterdam, Berlin, Bern, Rome, Stockholm, ViennaSelect other time zone

The GMT offsets shown reflect the offsets at the moment of the conference.

Time Band

Display full programSpecify a time band

Save

Session Program

Sun 17 Jan
Displayed time zone: Amsterdam, Berlin, Bern, Rome, Stockholm, Vienna change

18:00 - 19:30	Session 2LAFI at LAFI Chair(s): Dougal Maclaurin Google Research

18:00 12m Talk		Enzyme: High-Performance Automatic Differentiation of LLVM LAFI William S. Moses Massachusetts Institute of Technology, Valentin Churavy MIT CSAIL
18:12 12m Talk		Parametric Inversion of Non-Invertible Programs LAFI Zenna Tavares Massachusetts Institute of Technology, Javier Burroni University of Massachusetts Amherst, Edgar Minasyan Princeton University, David Morejon Massachusetts Institute of Technology, Armando Solar-Lezama Massachusetts Institute of Technology File Attached
18:25 12m Talk		Bayesian Neural Ordinary Differential Equations LAFI Raj Dandekar MIT, Vaibhav Dixit Julia Computing, Mohamed Tarek UNSW Canberra, Australia, Aslan Garcia Valadez National Autonomous University of Mexico, Chris Rackauckas MIT Pre-print Media Attached File Attached
18:38 12m Talk		On the Automatic Derivation of Importance Samplers from Pairs of Probabilistic Programs LAFI Alexander K. Lew Massachusetts Institute of Technology, USA, Ben Sherman , Marco Cusumano-Towner MIT-CSAIL, Michael Carbin Massachusetts Institute of Technology, Vikash K. Mansinghka MIT Media Attached
18:51 12m Talk		Decomposing reverse-mode automatic differentiation LAFI Roy Frostig Google Research, Matthew J. Johnson Google Brain, Dougal Maclaurin Google Research, Adam Paszke Google Research, Alexey Radul Google Research File Attached
19:04 12m Talk		Genify.jl: Transforming Julia into Gen to enable programmable inference LAFI Tan Zhi-Xuan Massachusetts Institute of Technology, McCoy R. Becker Charles River Analytics, Vikash K. Mansinghka MIT Media Attached File Attached
19:17 12m Talk		Probabilistic Inference Using Generators: the Statues Algorithm LAFI Pierre Denis independent scholar Link to publication DOI Pre-print Media Attached File Attached