18 papers
Sharp First-Order Lower Bounds under -Polyak-Lojasiewicz Conditions
Saeed Masiha, Negar Kiyavash, Patrick Thiran
We study first-order oracle complexity under the -Polyak-Lojasiewicz condition for . For , we first show that global -sm…
Active Context Selection Improves Simple Regret in Contextual Bandits
Mohammad Shahverdikondori, Jalal Etesami, Negar Kiyavash
We study the contextual multi-armed bandit problem with a finite context space (a.k.a. subpopulations), where the learner recommends a best action for each context and is evaluated…
Pure Exploration Beyond Reward Feedback: The Role of Post-Action Context
Mohammad Shahverdikondori, Amir Mohammad Abouei, Alireza Rezaeimoghadam +1
We introduce the problem of best arm identification (BAI) with post-action context, a new BAI problem in a stochastic multi-armed bandit environment and the fixed-confidence settin…
Select-then-differentiate: Solving Bilevel Optimization with Manifold Lower-level Solution Sets
Saeed Masiha, Zebang Shen, Negar Kiyavash +1
We study optimistic bilevel optimization when the lower-level problem has a non-isolated manifold of minimizers. In this setting, the hyper-objective may be non-differentiable beca…
Inference Time Causal Probing in LLMs
Sadegh Khorasani, Saber Salehkaleybar, Negar Kiyavash +1
Causal probing methods aim to test and control how internal representations influence the behavior of generative models. In causal probing, an intervention modifies hidden states s…
Graph Learning Is Suboptimal in Causal Bandits
Mohammad Shahverdikondori, Jalal Etesami, Negar Kiyavash
We study regret minimization in causal bandits under causal sufficiency where the underlying causal structure is not known to the agent. Previous work has focused on identifying th…