Coarsening Optimization for Differentiable Programming

no code implementations • 5 Oct 2021 • Xipeng Shen, Guoqiang Zhang, Irene Dea, Samantha Andow, Emilio Arroyo-Fang, Neal Gafter, Johann George, Melissa Grueter, Erik Meijer, Steffi Stumpos, Alanna Tempest, Christy Warden, Shannon Yang

This paper presents a novel optimization for differentiable programming named coarsening optimization.

Localized Uncertainty Attacks

no code implementations • 17 Jun 2021 • Ousmane Amadou Dia, Theofanis Karaletsos, Caner Hazirbas, Cristian Canton Ferrer, Ilknur Kaynar Kabul, Erik Meijer

Under this threat model, we create adversarial examples by perturbing only regions in the inputs where a classifier is uncertain.

Accelerating Metropolis-Hastings with Lightweight Inference Compilation

1 code implementation • 23 Oct 2020 • Feynman Liang, Nimar Arora, Nazanin Tehrani, Yucen Li, Michael Tingley, Erik Meijer

In order to construct accurate proposers for Metropolis-Hastings Markov Chain Monte Carlo, we integrate ideas from probabilistic graphical models and neural networks in an open-source framework we call Lightweight Inference Compilation (LIC).

Probabilistic Programming

PPL Bench: Evaluation Framework For Probabilistic Programming Languages

1 code implementation • 17 Oct 2020 • Sourabh Kulkarni, Kinjal Divesh Shah, Nimar Arora, Xiaoyan Wang, Yucen Lily Li, Nazanin Khosravani Tehrani, Michael Tingley, David Noursi, Narjes Torabi, Sepehr Akhavan Masouleh, Eric Lippert, Erik Meijer

The benchmark includes data generation and evaluation code for a number of models as well as implementations in some common PPLs.

Probabilistic Programming

Ownership at Large -- Open Problems and Challenges in Ownership Management

no code implementations • 15 Apr 2020 • John Ahlgren, Maria Eugenia Berezin, Kinga Bojarczuk, Elena Dulskyte, Inna Dvortsova, Johann George, Natalija Gucevska, Mark Harman, Shan He, Ralf Lämmel, Erik Meijer, Silvia Sapora, Justin Spahr-Summers

Software-intensive organizations rely on large numbers of software assets of different types, e. g., source-code files, tables in the data warehouse, and software configurations.

BIG-bench Machine Learning Management

WES: Agent-based User Interaction Simulation on Real Infrastructure

no code implementations • 11 Apr 2020 • John Ahlgren, Maria Eugenia Berezin, Kinga Bojarczuk, Elena Dulskyte, Inna Dvortsova, Johann George, Natalija Gucevska, Mark Harman, Ralf Lämmel, Erik Meijer, Silvia Sapora, Justin Spahr-Summers

We introduce the Web-Enabled Simulation (WES) research agenda, and describe FACEBOOK's WW system.

BIG-bench Machine Learning

Newtonian Monte Carlo: single-site MCMC meets second-order gradient methods

2 code implementations • 15 Jan 2020 • Nimar S. Arora, Nazanin Khosravani Tehrani, Kinjal Divesh Shah, Michael Tingley, Yucen Lily Li, Narjes Torabi, David Noursi, Sepehr Akhavan Masouleh, Eric Lippert, Erik Meijer

NMC is similar to the Newton-Raphson update in optimization where the second order gradient is used to automatically scale the step size in each dimension.

Second-order methods Variational Inference

Newtonian Monte Carlo: a second-order gradient method for speeding up MCMC

no code implementations • pproximateinference AABI Symposium 2019 • Nimar S. Arora, Nazanin Khosravani Tehrani, Kinjal Divesh Shah, Michael Tingley, Yucen Lily Li, Narjes Torabi, David Noursi, Sepehr Akhavan Masouleh, Eric Lippert, Erik Meijer

NMC is similar to the Newton-Raphson update in optimization where the second order gradient is used to automatically scale the step size in each dimension.

Variational Inference

Gradient Descent: The Ultimate Optimizer

2 code implementations • 29 Sep 2019 • Kartik Chandra, Audrey Xie, Jonathan Ragan-Kelley, Erik Meijer

This allows us to easily apply the method to other optimizers and hyperparameters (e. g. momentum coefficients).

BIG-bench Machine Learning Hyperparameter Optimization

MLSys: The New Frontier of Machine Learning Systems

no code implementations • 29 Mar 2019 • Alexander Ratner, Dan Alistarh, Gustavo Alonso, David G. Andersen, Peter Bailis, Sarah Bird, Nicholas Carlini, Bryan Catanzaro, Jennifer Chayes, Eric Chung, Bill Dally, Jeff Dean, Inderjit S. Dhillon, Alexandros Dimakis, Pradeep Dubey, Charles Elkan, Grigori Fursin, Gregory R. Ganger, Lise Getoor, Phillip B. Gibbons, Garth A. Gibson, Joseph E. Gonzalez, Justin Gottschlich, Song Han, Kim Hazelwood, Furong Huang, Martin Jaggi, Kevin Jamieson, Michael. I. Jordan, Gauri Joshi, Rania Khalaf, Jason Knight, Jakub Konečný, Tim Kraska, Arun Kumar, Anastasios Kyrillidis, Aparna Lakshmiratan, Jing Li, Samuel Madden, H. Brendan McMahan, Erik Meijer, Ioannis Mitliagkas, Rajat Monga, Derek Murray, Kunle Olukotun, Dimitris Papailiopoulos, Gennady Pekhimenko, Theodoros Rekatsinas, Afshin Rostamizadeh, Christopher Ré, Christopher De Sa, Hanie Sedghi, Siddhartha Sen, Virginia Smith, Alex Smola, Dawn Song, Evan Sparks, Ion Stoica, Vivienne Sze, Madeleine Udell, Joaquin Vanschoren, Shivaram Venkataraman, Rashmi Vinayak, Markus Weimer, Andrew Gordon Wilson, Eric Xing, Matei Zaharia, Ce Zhang, Ameet Talwalkar

Machine learning (ML) techniques are enjoying rapidly increasing adoption.

BIG-bench Machine Learning