Hello there!

My name is Luiz Chamon. I study the theoretical underpinnings of constrained learning and its applications in signal processing, control, and machine learning.

I am currently on the academic job market. You can find a 5-minute video pitch of my research plan and more information in my virtual application package.

News

• April 1st, 2021 – I will be giving a seminar at Microsoft Research.

• March 23rd, 2021 – "Approximately supermodular scheduling subject to matroid constraints" was accepted for publication on IEEE TAC (arXiv).

• March 10th, 2021 – New preprint: "Constrained learning with non-convex losses" (arXiv).

• February 26th, 2021 – New papers accepted at ACC 2021:

  • "Towards safe continuing task reinforcement learning" [paper]
  • "Source seeking in unknown environments with convex obstacles" [paper]

• February 25th, 2020 – New preprint: "State augmented constrained reinforcement learning: Overcoming the limitations of learning with rewards" (arXiv).

• February 1st, 2021 – I will be giving a seminar at the Massachusetts Institute of Technology (EECS).

• January 18th, 2021 – Check out our paper this week at EUSIPCO 2020: "Graphon filters: Signal processing in very large graphs" (Wednesday 9:10am, EST) [paper]

• January 8th, 2021 – I will be giving a seminar at the Johns Hopkins Mathematical Institute for Data Science (MINDS).

• January 6th, 2021 – I will be giving a seminar at the Toyota Technological Institute at Chicago (TTIC).

• December 30th, 2020 – "Approximate supermodularity of Kalman filter sensor selection" published on IEEE TAC (paper, IEEEXplore).

• December 15th, 2020 – Check out my papers at IEEE CDC this week:

  • "Risk-constrained linear quadratic regulators" (Wednesday 9:15am, EST) [paper]
  • "Resilient control: Compromising to adapt" (Friday 8:15am, EST) [paper]

• December 5th, 2020 – Check out my papers at NeurIPS next week:

  • "Probably Approximately Correct Constrained Learning" (Tuesday 9–11am, PST)
  • "Transferability of Graphon Neural Networks" (Thursday 9–11am, PST)

• December 3rd, 2020 – I've released the first version of csl, a python package for constrained learning. You can get the package on GitHub and check out applications in fairness and robustness in the documentations.

• October 13th, 2020 – "Probably Approximately Correct Constrained Learning" (arXiv) and "Transferability of Graphon Neural Networks" (arXiv) accepted at NeurIPS 2020.

• September 5th, 2020 – "Resilient control: Compromising to adapt" (arXiv) and "Risk-constrained linear-quadratic regulators" (arXiv) accepted at IEEE CDC 2020.

• May 5th, 2020 – I received the best student paper and best paper awards at ICASSP 2020. Check out papers [1, 2] and video [1].

• April 1st, 2020 – New preprint: "Graphon signal processing" submitted to IEEE TSP (arXiv).

• March 20th, 2020 – "Functional nonlinear sparse models" was accepted for publication on IEEE TSP (arXiv).

• February 7th, 2020 – Patent "Sparse cascaded-integrator-comb filters" with Analog Devices is out.