no code implementations • 18 Mar 2024 • Juan Elenter, Luiz F. O. Chamon, Alejandro Ribeiro
These requirements can be imposed (with generalization guarantees) by formulating constrained learning problems that can then be tackled by dual ascent algorithms.
no code implementations • 5 Jan 2024 • Luana Ruiz, Luiz F. O. Chamon, Alejandro Ribeiro
This technical note addresses an issue [arXiv:2310. 14683] with the proof (but not the statement) of [arXiv:2003. 05030, Proposition 4].
no code implementations • NeurIPS 2023 • Ignacio Hounie, Alejandro Ribeiro, Luiz F. O. Chamon
This paper presents a constrained learning approach that adapts the requirements while simultaneously solving the learning task.
no code implementations • 1 Oct 2022 • Juan Cervino, Luiz F. O. Chamon, Benjamin D. Haeffele, Rene Vidal, Alejandro Ribeiro
To do so, it shows that under typical conditions the problem of learning a Lipschitz continuous function on a manifold is equivalent to a dynamically weighted manifold regularization problem.
1 code implementation • 29 Sep 2022 • Ignacio Hounie, Luiz F. O. Chamon, Alejandro Ribeiro
Despite its ubiquity, its effectiveness depends on the choices of which transformations to apply, when to do so, and how often.
Ranked #1 on Image Classification on SVHN (Percentage correct metric)
1 code implementation • 2 Feb 2022 • Alexander Robey, Luiz F. O. Chamon, George J. Pappas, Hamed Hassani
From a theoretical point of view, this framework overcomes the trade-offs between the performance and the sample-complexity of worst-case and average-case learning.
no code implementations • 9 Dec 2021 • Luana Ruiz, Luiz F. O. Chamon, Alejandro Ribeiro
In this paper, we study the problem of training GNNs on graphs of moderate size and transferring them to large-scale graphs.
no code implementations • NeurIPS 2021 • Alexander Robey, Luiz F. O. Chamon, George J. Pappas, Hamed Hassani, Alejandro Ribeiro
In particular, we leverage semi-infinite optimization and non-convex duality theory to show that adversarial training is equivalent to a statistical problem over perturbation distributions, which we characterize completely.
no code implementations • 8 Mar 2021 • Luiz F. O. Chamon, Santiago Paternain, Miguel Calvo-Fullana, Alejandro Ribeiro
In this paper, we overcome this issue by learning in the empirical dual domain, where constrained statistical learning problems become unconstrained and deterministic.
no code implementations • 24 Feb 2021 • Miguel Calvo-Fullana, Luiz F. O. Chamon, Santiago Paternain
However, to transfer from learning safety to learning safely, there are two hurdles that need to be overcome: (i) it has to be possible to learn the policy without having to re-initialize the system; and (ii) the rollouts of the system need to be in themselves safe.
no code implementations • 23 Feb 2021 • Miguel Calvo-Fullana, Santiago Paternain, Luiz F. O. Chamon, Alejandro Ribeiro
Thus, as we illustrate by an example, while previous methods can fail at finding optimal policies, running the dual dynamics while executing the augmented policy yields an algorithm that provably samples actions from the optimal policy.
no code implementations • 24 Nov 2020 • Luiz F. O. Chamon, Santiago Paternain, Alejandro Ribeiro
Prediction credibility measures, in the form of confidence intervals or probability distributions, are fundamental in statistics and machine learning to characterize model robustness, detect out-of-distribution samples (outliers), and protect against adversarial attacks.
no code implementations • 10 Mar 2020 • Luana Ruiz, Luiz F. O. Chamon, Alejandro Ribeiro
Graphons are infinite-dimensional objects that represent the limit of convergent sequences of graphs as their number of nodes goes to infinity.