14 papers
To Defer or Not to Defer? Online Learning with LLMs and Congested Experts
Vasileios Saketos, William Reveillard, Alexandre Proutiere +2
Modern AI systems increasingly combine large language models (LLMs) with human experts, using the LLM to handle routine queries while deferring uncertain cases for expert review. W…
Instance-Optimal Estimation with Multiple LLM Judges on a Budget
Junghyun Lee, Sanghwa Kim, Yassir Jedra +2
Evaluating large language models increasingly relies on LLM-as-a-judge protocols, but such evaluations remain costly: different judges have different prices and reliabilities, and…
Optimal Centered Active Excitation in Linear System Identification
Kaito Ito, Alexandre Proutiere
We propose an active learning algorithm for linear system identification with optimal centered noise excitation. Notably, our algorithm, based on ordinary least squares and semidef…
Curvature-Guided LoRA: Matching Full Fine-Tuning in Function Space
Frédéric Zheng, Alexandre Proutière
Parameter-efficient fine-tuning methods such as LoRA enable efficient adaptation of large pretrained models, but often lag behind full fine-tuning in both convergence speed and fin…
Minimizing Human Intervention in Online Classification
William Réveillard, Vasileios Saketos, Alexandre Proutiere +1
Training or fine-tuning large language model (LLM)-based systems often requires costly human feedback, yet there is limited understanding of how to minimize such intervention while…
Near-optimal Rank Adaptive Inference of High Dimensional Matrices
Frédéric Zheng, Yassir Jedra, Alexandre Proutiere
We address the problem of estimating a high-dimensional matrix from linear measurements, with a focus on designing optimal rank-adaptive algorithms. These algorithms infer the matr…
Shift Before You Learn: Enabling Low-Rank Representations in Reinforcement Learning
Bastien Dubail, Stefan Stojanovic, Alexandre Proutière
Low-rank structure is a common implicit assumption in many modern reinforcement learning (RL) algorithms. For instance, reward-free and goal-conditioned RL methods often presume th…
Minimal Order Recovery through Rank-adaptive Identification
Frédéric Zheng, Yassir Jedra, Alexandre Proutière
This paper addresses the problem of identifying linear systems from noisy input-output trajectories. We introduce Thresholded Ho-Kalman, an algorithm that leverages a rank-adaptive…
Near-Optimal Clustering in Mixture of Markov Chains
Junghyun Lee, Yassir Jedra, Alexandre Proutière +1
We study the problem of clustering trajectories of length , each generated by one of K unknown ergodic Markov chains over a finite state space of size . We derive an inst…
Model-free Low-Rank Reinforcement Learning via Leveraged Entry-wise Matrix Estimation
Stefan Stojanovic, Yassir Jedra, Alexandre Proutiere
We consider the problem of learning an -optimal policy in controlled dynamical systems with low-rank latent structure. For this problem, we present LoRa-PI (Low-Rank P…
Conformal Predictions under Markovian Data
Frédéric Zheng, Alexandre Proutiere
We study the split Conformal Prediction method when applied to Markovian data. We quantify the gap in terms of coverage induced by the correlations in the data (compared to exchang…
Low-Rank Bandits via Tight Two-to-Infinity Singular Subspace Recovery
Yassir Jedra, William Réveillard, Stefan Stojanovic +1
We study contextual bandits with low-rank structure where, in each round, if the (context, arm) pair is selected, the learner observes a noisy sample of th…
Spectral Entry-wise Matrix Estimation for Low-Rank Reinforcement Learning
Stefan Stojanovic, Yassir Jedra, Alexandre Proutiere
We study matrix estimation problems arising in reinforcement learning (RL) with low-rank structure. In low-rank bandits, the matrix to be recovered specifies the expected arm rewar…
Sub-linear Regret in Adaptive Model Predictive Control
Damianos Tranos, Alexandre Proutiere
We consider the problem of adaptive Model Predictive Control (MPC) for uncertain linear-systems with additive disturbances and with state and input constraints. We present STT-MPC…