40 citations · 151 across the 17 of their papers we have counts for
12 papers · 1 filter
Towards Faithful Data Selection: Statistical Limits and Scalable Algorithms
Ahmed Mehdi Inane, Mahdi Haghifam, Gintare Karolina Dziugaite +1
What does it mean to have an optimal data selection algorithm? Optimality is conventionally framed as finding a subset that minimizes the population error of the classifier trained…
Adaptive Generate-Rank-Verify: Inference-Time Search with Costly Verification
Shaddin Dughmi, Mahdi Haghifam, Yusuf Hakan Kalayci
Many inference-time language-model pipelines combine a cheap reward signal with an expensive verifier, such as exact answer checking in mathematical reasoning or hidden-test execut…
The Distillation Game: Adaptive Evaluations & Efficient Defenses
Youssef Allouah, Mahdi Haghifam, Sanmi Koyejo +1
Distillation attacks create a deployment trade-off for model providers: the same outputs that make a model more useful can also make it easier to imitate. We study this trade-off t…
The Sample Complexity of Membership Inference and Privacy Auditing
Mahdi Haghifam, Adam Smith, Jonathan Ullman
A membership-inference attack gets the output of a learning algorithm, and a target individual, and tries to determine whether this individual is a member of the training data or a…
On Traceability in Stochastic Convex Optimization
Sasha Voitovych, Mahdi Haghifam, Idan Attias +3
In this paper, we investigate the necessity of traceability for accurate learning in stochastic convex optimization (SCO) under geometries. Informally, we say a learning a…
Private Geometric Median
Mahdi Haghifam, Thomas Steinke, Jonathan Ullman
In this paper, we study differentially private (DP) algorithms for computing the geometric median (GM) of a dataset: Given points, in , the goal i…
Information Complexity of Stochastic Convex Optimization: Applications to Generalization and Memorization
Idan Attias, Gintare Karolina Dziugaite, Mahdi Haghifam +2
In this work, we investigate the interplay between memorization and learning in the context of \emph{stochastic convex optimization} (SCO). We define memorization via the informati…
Faster Differentially Private Convex Optimization via Second-Order Methods
Arun Ganesh, Mahdi Haghifam, Thomas Steinke +1
Differentially private (stochastic) gradient descent is the workhorse of DP private machine learning in both the convex and non-convex settings. Without privacy constraints, second…
Why Is Public Pretraining Necessary for Private Model Training?
Arun Ganesh, Mahdi Haghifam, Milad Nasr +5
In the privacy-utility tradeoff of a model trained on benchmark language and vision tasks, remarkable improvements have been widely reported with the use of pretraining on publicly…
Limitations of Information-Theoretic Generalization Bounds for Gradient Descent Methods in Stochastic Convex Optimization
Mahdi Haghifam, Borja Rodríguez-Gálvez, Ragnar Thobaben +3
To date, no "information-theoretic" frameworks for reasoning about generalization error have been shown to establish minimax rates for gradient descent in the setting of stochastic…
Understanding Generalization via Leave-One-Out Conditional Mutual Information
Mahdi Haghifam, Shay Moran, Daniel M. Roy +1
We study the mutual information between (certain summaries of) the output of a learning algorithm and its training data, conditional on a supersample of i.i.d. data from…
On the Information Complexity of Proper Learners for VC Classes in the Realizable Case
Mahdi Haghifam, Gintare Karolina Dziugaite, Shay Moran +1
We provide a negative resolution to a conjecture of Steinke and Zakynthinou (2020a), by showing that their bound on the conditional mutual information (CMI) of proper learners of V…