Online Infinite-Dimensional Regression: Learning Linear Operators

We consider the problem of learning linear operators under squared loss between two infinite-dimensional Hilbert spaces in the online setting. We show that the class of linear operators with uniformly bounded $p$-Schatten norm is online learnable for any $p \in [1, \infty)$. On the other hand, we prove an impossibility result by showing that the class of uniformly bounded linear operators with respect to the operator norm is \textit{not} online learnable. Moreover, we show a separation between online uniform convergence and online learnability by identifying a class of bounded linear operators that is online learnable but uniform convergence does not hold. Finally, we prove that the impossibility result and the separation between uniform convergence and learnability also hold in the agnostic PAC setting.

On the Minimax Regret in Online Ranking with Top-k Feedback

In online ranking, a learning algorithm sequentially ranks a set of items and receives feedback on its ranking in the form of relevance scores. Since obtaining relevance scores typically involves human annotation, it is of great interest to consider a partial feedback setting where feedback is restricted to the top-$k$ items in the rankings. Chaudhuri and Tewari [2017] developed a framework to analyze online ranking algorithms with top $k$ feedback. A key element in their work was the use of techniques from partial monitoring. In this paper, we further investigate online ranking with top $k$ feedback and solve some open problems posed by Chaudhuri and Tewari [2017]. We provide a full characterization of minimax regret rates with the top $k$ feedback model for all $k$ and for the following ranking performance measures: Pairwise Loss, Discounted Cumulative Gain, and Precision@n. In addition, we give an efficient algorithm that achieves the minimax regret rate for Precision@n.

Multiclass Online Learnability under Bandit Feedback

We study online multiclass classification under bandit feedback. We extend the results of (daniely2013price) by showing that the finiteness of the Bandit Littlestone dimension is necessary and sufficient for bandit online multiclass learnability even when the label space is unbounded. Our result complements the recent work by (hanneke2023multiclass) who show that the Littlestone dimension characterizes online multiclass learnability in the full-information setting when the label space is unbounded.

A Combinatorial Characterization of Online Learning Games with Bounded Losses

We study the online learnability of hypothesis classes with respect to arbitrary, but bounded, loss functions. We give a new scale-sensitive combinatorial dimension, named the sequential Minimax dimension, and show that it gives a tight quantitative characterization of online learnability. As applications, we give the first quantitative characterization of online learnability for two natural learning settings: vector-valued regression and multilabel classification.

Online Learning with Set-Valued Feedback

We study a variant of online multiclass classification where the learner predicts a single label but receives a \textit{set of labels} as feedback. In this model, the learner is penalized for not outputting a label contained in the revealed set. We show that unlike online multiclass learning with single-label feedback, deterministic and randomized online learnability are \textit{not equivalent} even in the realizable setting with set-valued feedback. Accordingly, we give two new combinatorial dimensions, named the Set Littlestone and Measure Shattering dimension, that tightly characterize deterministic and randomized online learnability respectively in the realizable setting. In addition, we show that the Measure Shattering dimension tightly characterizes online learnability in the agnostic setting. Finally, we show that practical learning settings like online multilabel ranking, online multilabel classification, and online interval learning are specific instances of our general framework.

A Primal-Dual-Critic Algorithm for Offline Constrained Reinforcement Learning

Offline constrained reinforcement learning (RL) aims to learn a policy that maximizes the expected cumulative reward subject to constraints on expected value of cost functions using an existing dataset. In this paper, we propose Primal-Dual-Critic Algorithm (PDCA), a novel algorithm for offline constrained RL with general function approximation. PDCA runs a primal-dual algorithm on the Lagrangian function estimated by critics. The primal player employs a no-regret policy optimization oracle to maximize the Lagrangian estimate given any choices of the critics and the dual player. The dual player employs a no-regret online linear optimization oracle to minimize the Lagrangian estimate given any choices of the critics and the primal player. We show that PDCA can successfully find a near saddle point of the Lagrangian, which is nearly optimal for the constrained RL problem. Unlike previous work that requires concentrability and strong Bellman completeness assumptions, PDCA only requires concentrability and value function/marginalized importance weight realizability assumptions.

Variational Inference with Coverage Guarantees

Amortized variational inference produces a posterior approximator that can compute a posterior approximation given any new observation. Unfortunately, there are few guarantees about the quality of these approximate posteriors. We propose Conformalized Amortized Neural Variational Inference (CANVI), a procedure that is scalable, easily implemented, and provides guaranteed marginal coverage. Given a collection of candidate amortized posterior approximators, CANVI constructs conformalized predictors based on each candidate, compares the predictors using a metric known as predictive efficiency, and returns the most efficient predictor. CANVI ensures that the resulting predictor constructs regions that contain the truth with high probability (exactly how high is prespecified by the user). CANVI is agnostic to design decisions in formulating the candidate approximators and only requires access to samples from the forward model, permitting its use in likelihood-free settings. We prove lower bounds on the predictive efficiency of the regions produced by CANVI and explore how the quality of a posterior approximation relates to the predictive efficiency of prediction regions based on that approximation. Finally, we demonstrate the accurate calibration and high predictive efficiency of CANVI on a suite of simulation-based inference benchmark tasks and an important scientific task: analyzing galaxy emission spectra.

On the Learnability of Multilabel Ranking

Multilabel ranking is a central task in machine learning with widespread applications to web search, news stories, recommender systems, etc. However, the most fundamental question of learnability in a multilabel ranking setting remains unanswered. In this paper, we characterize the learnability of multilabel ranking problems in both the batch and online settings for a large family of ranking losses. Along the way, we also give the first equivalence class of ranking losses based on learnability.

A Characterization of Online Multiclass Learnability

We consider the problem of online multiclass learning when the number of labels is unbounded. We show that the Multiclass Littlestone dimension, first introduced in \cite{DanielyERMprinciple}, continues to characterize online learnability in this setting. Our result complements the recent work by \cite{Brukhimetal2022} who give a characterization of batch multiclass learnability when the label space is unbounded.