Stephen Becker

About

Stephen Becker is an Associate Professor and the head of the Becker research group at the University of Colorado Boulder, having joined the university in 2014. His research group focuses on a variety of topics including optimization, quantum tomography, equation discovery, scientific data compression, and machine learning methods. The group works on advanced mathematical and computational techniques such as one-pass compression, zeroth order optimization, Gauss-Newton methods for quantum tomography maximum likelihood estimation, and non-uniform convolutions for neural networks. Becker's lab collaborates with students and postdoctoral researchers across multiple disciplines including computer science, electrical, computer, and energy engineering, physics, and applied mathematics. The group also engages with many PhD students from other research groups, indicating a broad interdisciplinary approach to their research endeavors.

Research topics

• Developmental psychology
• Clinical psychology
• Neuroscience
• Psychology

Selected publications

• Bi-fidelity variational auto-encoder for uncertainty quantification

  Computer Methods in Applied Mechanics and Engineering · 2024-02-13 · 16 citations

  articleOpen access
  PublisherOA PDFDOI

• Theory of versatile fidelity estimation with confidence

  Physical review. A/Physical review, A · 2024-07-10 · 8 citations

  articleSenior author

  Estimating the fidelity with a target state is important in quantum information tasks. Many fidelity-estimation techniques present a suitable measurement scheme to perform the estimation. In contrast, we present techniques that allow the experimentalist to choose a convenient measurement setting. Our primary focus lies on a method that constructs an estimator with nearly minimax optimal confidence intervals for any specified measurement setting. We demonstrate, through a combination of theoretic…

  PublisherDOI

• Versatile Fidelity Estimation with Confidence

  Physical Review Letters · 2024-07-10 · 7 citations

  articleSenior author

  As quantum devices become more complex and the requirements on these devices become more demanding, it is crucial to be able to verify the performance of such devices in a scalable and reliable fashion. A cornerstone task in this challenge is quantifying how close an experimentally prepared quantum state is to the desired one. Here we present a method to construct an estimator for the quantum state fidelity that is compatible with any measurement protocol. Our method provides a confidence interv…

  PublisherDOI

• QuadConv: Quadrature-based convolutions with applications to non-uniform PDE data compression

  Journal of Computational Physics · 2023-11-20 · 3 citations

  article
  PublisherDOI

• Aligning to What? Limits to RLHF Based Alignment

  2025-01-01 · 2 citations

  articleOpen access

  Reinforcement Learning from Human Feedback (RLHF) is increasingly used to align large language models (LLMs) with human preferences.However, the effectiveness of RLHF in addressing underlying biases remains unclear.This study investigates the relationship between RLHF and both covert and overt biases in LLMs, particularly focusing on biases against African Americans.We applied various RLHF techniques (DPO, ORPO, and RLOO) to Llama 3 8B and evaluated the covert and overt biases of the resulting m…

  PublisherOA PDFDOI

Recent grants

• AMPS: Online and Model-Free Optimization of Power and Energy Systems

  NSF · $357k · 2019–2024

• Direct Estimates and Confidence Intervals for Fidelity of Quantum States

  NSF · $249k · 2021–2025

• Extraction of Information from Scientific Simulations

  NSF · $150k · 2018–2022

Frequent coauthors

• Alireza Doostan

  University of Colorado Boulder

  30 shared

• Liam Madden

  24 shared

• Farhad Pourkamali‐Anaraki

  23 shared

• Emiliano Dall’Anese

  University of Colorado Boulder

  23 shared

• Volkan Cevher

  22 shared

• Osman Asif Malik

  Lawrence Berkeley National Laboratory

  18 shared

• Akshay Seshadri

  University of Colorado System

  17 shared

• Thomas Monz

  Alpine Quantum Technologies (Austria)

  16 shared

Education

• Ph.D.

  California Institute of Technology

• Other

  Fondation Sciences Mathématiques de Paris at Paris 6 (JLL lab)

• Other, Mathematical Sciences

  IBM Research in Yorktown Heights, NY

Awards & honors

• Beal-Orchard-Hays prize (2015)

Similar researchers at University of Colorado Boulder

• Mark J. Ablowitzh-86
• James Curryh-64
• Gregory Beylkinh-47