Navigating Ensemble Configurations for Algorithmic Fairness

no code implementations • 11 Oct 2022 • Michael Feffer, Martin Hirzel, Samuel C. Hoffman, Kiran Kate, Parikshit Ram, Avraham Shinnar

Bias mitigators can improve algorithmic fairness in machine learning models, but their effect on fairness is often not stable across data splits.

Ensemble Learning Fairness +1

Formalization of a Stochastic Approximation Theorem

1 code implementation • 12 Feb 2022 • Koundinya Vajjha, Barry Trager, Avraham Shinnar, Vasily Pestun

Stochastic approximation algorithms are iterative procedures which are used to approximate a target value in an environment where the target is unknown and direct observations are corrupted by noise.

An Empirical Study of Modular Bias Mitigators and Ensembles

no code implementations • 1 Feb 2022 • Michael Feffer, Martin Hirzel, Samuel C. Hoffman, Kiran Kate, Parikshit Ram, Avraham Shinnar

A popular approach to train more stable models is ensemble learning.

Ensemble Learning Fairness

Pipeline Combinators for Gradual AutoML

no code implementations • NeurIPS 2021 • Guillaume Baudart, Martin Hirzel, Kiran Kate, Parikshit Ram, Avraham Shinnar, Jason Tsay

Automated machine learning (AutoML) can make data scientists more productive.

AutoML BIG-bench Machine Learning +1

RASL: Relational Algebra in Scikit-Learn Pipelines

no code implementations • NeurIPS Workshop DBAI 2021 • Chirag Sahni, Kiran Kate, Avraham Shinnar, Hoang Thanh Lam, Martin Hirzel

Integrating data preparation with machine-learning (ML) pipelines has been a long- standing challenge.

BIG-bench Machine Learning

CertRL: Formalizing Convergence Proofs for Value and Policy Iteration in Coq

1 code implementation • 23 Sep 2020 • Koundinya Vajjha, Avraham Shinnar, Vasily Pestun, Barry Trager, Nathan Fulton

Reinforcement learning algorithms solve sequential decision-making problems in probabilistic environments by optimizing for long-term reward.

Decision Making reinforcement-learning +1

Lale: Consistent Automated Machine Learning

1 code implementation • 4 Jul 2020 • Guillaume Baudart, Martin Hirzel, Kiran Kate, Parikshit Ram, Avraham Shinnar

Automated machine learning makes it easier for data scientists to develop pipelines by searching over possible choices for hyperparameters, algorithms, and even pipeline topologies.

BIG-bench Machine Learning

Type-Driven Automated Learning with Lale

2 code implementations • 24 May 2019 • Martin Hirzel, Kiran Kate, Avraham Shinnar, Subhrajit Roy, Parikshit Ram

Machine-learning automation tools, ranging from humble grid-search to hyperopt, auto-sklearn, and TPOT, help explore large search spaces of possible pipelines.

Time Series Time Series Analysis +1

Yaps: Python Frontend to Stan

1 code implementation • 6 Dec 2018 • Guillaume Baudart, Martin Hirzel, Kiran Kate, Louis Mandel, Avraham Shinnar

Stan is a popular probabilistic programming language with a self-contained syntax and semantics that is close to graphical models.

Programming Languages

Compiling Stan to Generative Probabilistic Languages and Extension to Deep Probabilistic Programming

1 code implementation • 30 Sep 2018 • Guillaume Baudart, Javier Burroni, Martin Hirzel, Louis Mandel, Avraham Shinnar

We use our compilation scheme to build two new backends for the Stanc3 compiler targeting Pyro and NumPyro.

Probabilistic Programming Representation Learning +1

Ariadne: Analysis for Machine Learning Program

no code implementations • 10 May 2018 • Julian Dolby, Avraham Shinnar, Allison Allain, Jenna Reinen

We report on Ariadne: applying a static framework, WALA, to machine learning code that uses TensorFlow.

Programming Languages