Statistical indistinguishability of learning algorithms
When two different parties use the same learning rule on their own data, how can we test
whether the distributions of the two outcomes are similar? In this paper, we study the …
whether the distributions of the two outcomes are similar? In this paper, we study the …
Replicable reinforcement learning
The replicability crisis in the social, behavioral, and data sciences has led to the formulation
of algorithm frameworks for replicability---ie, a requirement that an algorithm produce …
of algorithm frameworks for replicability---ie, a requirement that an algorithm produce …
Replicable clustering
We design replicable algorithms in the context of statistical clustering under the recently
introduced notion of replicability from Impagliazzo et al.[2022]. According to this definition, a …
introduced notion of replicability from Impagliazzo et al.[2022]. According to this definition, a …
Replicability in reinforcement learning
We initiate the mathematical study of replicability as an algorithmic property in the context of
reinforcement learning (RL). We focus on the fundamental setting of discounted tabular …
reinforcement learning (RL). We focus on the fundamental setting of discounted tabular …
List and certificate complexities in replicable learning
We investigate replicable learning algorithms. Informally a learning algorithm is replicable if
the algorithm outputs the same canonical hypothesis over multiple runs with high probability …
the algorithm outputs the same canonical hypothesis over multiple runs with high probability …
Replicable learning of large-margin halfspaces
We provide efficient replicable algorithms for the problem of learning large-margin
halfspaces. Our results improve upon the algorithms provided by Impagliazzo, Lei, Pitassi …
halfspaces. Our results improve upon the algorithms provided by Impagliazzo, Lei, Pitassi …
Optimal guarantees for algorithmic reproducibility and gradient complexity in convex optimization
Algorithmic reproducibility measures the deviation in outputs of machine learning algorithms
upon minor changes in the training process. Previous work suggests that first-order methods …
upon minor changes in the training process. Previous work suggests that first-order methods …
Can Probabilistic Feedback Drive User Impacts in Online Platforms?
A common explanation for negative user impacts of content recommender systems is
misalignment between the platform's objective and user welfare. In this work, we show that …
misalignment between the platform's objective and user welfare. In this work, we show that …
On the Computational Landscape of Replicable Learning
We study computational aspects of algorithmic replicability, a notion of stability introduced by
Impagliazzo, Lei, Pitassi, and Sorrell [2022]. Motivated by a recent line of work that …
Impagliazzo, Lei, Pitassi, and Sorrell [2022]. Motivated by a recent line of work that …
Replicability and stability in learning
Z Chase, S Moran, A Yehudayoff - arXiv preprint arXiv:2304.03757, 2023 - arxiv.org
Replicability is essential in science as it allows us to validate and verify research findings.
Impagliazzo, Lei, Pitassi and Sorrell (22) recently initiated the study of replicability in …
Impagliazzo, Lei, Pitassi and Sorrell (22) recently initiated the study of replicability in …