Near-Optimal Algorithms for Omniprediction

• URL: http://arxiv.org/abs/2501.17205v2
• Date: Thu, 30 Jan 2025 01:45:23 GMT
• Title: Near-Optimal Algorithms for Omniprediction
• Authors: Princewill Okoroafor, Robert Kleinberg, Michael P. Kim,
• Abstract summary: We give near-optimal learning algorithms for omniprediction, in both the online and offline settings.<n>Online learning algorithm achieves near-optimal complexity across a number of measures.<n> offline learning algorithm returns an efficient $(mathcalL_mathrmBV,mathcalH,varepsilon(m)$)
• Score: 6.874077229518565
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Omnipredictors are simple prediction functions that encode loss-minimizing predictions with respect to a hypothesis class $\mathcal{H}$, simultaneously for every loss function within a class of losses $\mathcal{L}$. In this work, we give near-optimal learning algorithms for omniprediction, in both the online and offline settings. To begin, we give an oracle-efficient online learning algorithm that acheives $(\mathcal{L},\mathcal{H})$-omniprediction with $\tilde{O}(\sqrt{T \log |\mathcal{H}|})$ regret for any class of Lipschitz loss functions $\mathcal{L} \subseteq \mathcal{L}_\mathrm{Lip}$. Quite surprisingly, this regret bound matches the optimal regret for \emph{minimization of a single loss function} (up to a $\sqrt{\log(T)}$ factor). Given this online algorithm, we develop an online-to-offline conversion that achieves near-optimal complexity across a number of measures. In particular, for all bounded loss functions within the class of Bounded Variation losses $\mathcal{L}_\mathrm{BV}$ (which include all convex, all Lipschitz, and all proper losses) and any (possibly-infinite) $\mathcal{H}$, we obtain an offline learning algorithm that, leveraging an (offline) ERM oracle and $m$ samples from $\mathcal{D}$, returns an efficient $(\mathcal{L}_{\mathrm{BV}},\mathcal{H},\varepsilon(m))$-omnipredictor for $\varepsilon(m)$ scaling near-linearly in the Rademacher complexity of $\mathrm{Th} \circ \mathcal{H}$.

Related papers

• Learning and Computation of $Φ$-Equilibria at the Frontier of Tractability [85.07238533644636]
  $Phi$-equilibria is a powerful and flexible framework at the heart of online learning and game theory. We show that an efficient online algorithm incurs average $Phi$-regret at most $epsilon$ using $textpoly(d, k)/epsilon2$ rounds. We also show nearly matching lower bounds in the online setting, thereby obtaining for the first time a family of deviations that captures the learnability of $Phi$-regret.
  arXiv  Detail & Related papers  (2025-02-25T19:08:26Z)
• Full Swap Regret and Discretized Calibration [18.944031222413294]
  We study the problem of minimizing swap regret in structured normal-form games. We introduce a new online learning problem we call emphfull swap regret minimization We also apply these tools to the problem of online forecasting to calibration error.
  arXiv  Detail & Related papers  (2025-02-13T13:49:52Z)
• Optimal and Efficient Algorithms for Decentralized Online Convex Optimization [51.00357162913229]
  Decentralized online convex optimization (D-OCO) is designed to minimize a sequence of global loss functions using only local computations and communications.<n>We develop a novel D-OCO algorithm that can reduce the regret bounds for convex and strongly convex functions to $tildeO(nrho-1/4sqrtT)$ and $tildeO(nrho-1/2log T)$.<n>Our analysis reveals that the projection-free variant can achieve $O(nT3/4)$ and $O(n
  arXiv  Detail & Related papers  (2024-02-14T13:44:16Z)
• Fast $(1+\varepsilon)$-Approximation Algorithms for Binary Matrix Factorization [54.29685789885059]
  We introduce efficient $(1+varepsilon)$-approximation algorithms for the binary matrix factorization (BMF) problem. The goal is to approximate $mathbfA$ as a product of low-rank factors. Our techniques generalize to other common variants of the BMF problem.
  arXiv  Detail & Related papers  (2023-06-02T18:55:27Z)
• Efficient Rate Optimal Regret for Adversarial Contextual MDPs Using Online Function Approximation [47.18926328995424]
  We present the OMG-CMDP! algorithm for regret minimization in adversarial Contextual MDPs. Our algorithm is efficient (assuming efficient online regression oracles) and simple and robust to approximation errors.
  arXiv  Detail & Related papers  (2023-03-02T18:27:00Z)
• Eluder-based Regret for Stochastic Contextual MDPs [43.19667415823089]
  We present the E-UC$3$RL algorithm for regret minimization in Contextual Markov Decision Processes (CMDPs) Our algorithm is efficient (assuming efficient offline regression oracles) and enjoys a regret guarantee of $ widetildeO(H3 sqrtT |S| |A|d_mathrmE(mathcalP)$.
  arXiv  Detail & Related papers  (2022-11-27T20:38:47Z)
• Private Isotonic Regression [54.32252900997422]
  We consider the problem of isotonic regression over a partially ordered set (poset) $mathcalX$ and for any Lipschitz loss function. We obtain a pure-DP algorithm that has an expected excess empirical risk of roughly $mathrmwidth(mathcalX) cdot log|mathcalX| / n$, where $mathrmwidth(mathcalX)$ is the width of the poset. We show that the bounds above are essentially the best that can be
  arXiv  Detail & Related papers  (2022-10-27T05:08:07Z)
• On Optimal Learning Under Targeted Data Poisoning [48.907813854832206]
  In this work we aim to characterize the smallest achievable error $epsilon=epsilon(eta)$ by the learner in the presence of such an adversary. Remarkably, we show that the upper bound can be attained by a deterministic learner.
  arXiv  Detail & Related papers  (2022-10-06T06:49:48Z)
• Online Learning with Bounded Recall [11.046741824529107]
  We study the problem of full-information online learning in the "bounded recall" setting popular in the study of repeated games. An online learning algorithm $mathcalA$ is $M$-$textitbounded-recall$ if its output at time $t$ can be written as a function of the $M$ previous rewards.
  arXiv  Detail & Related papers  (2022-05-28T20:52:52Z)
• Logarithmic Regret from Sublinear Hints [76.87432703516942]
  We show that an algorithm can obtain $O(log T)$ regret with just $O(sqrtT)$ hints under a natural query model. We also show that $o(sqrtT)$ hints cannot guarantee better than $Omega(sqrtT)$ regret.
  arXiv  Detail & Related papers  (2021-11-09T16:50:18Z)
• Online Sub-Sampling for Reinforcement Learning with General Function Approximation [111.01990889581243]
  In this paper, we establish an efficient online sub-sampling framework that measures the information gain of data points collected by an RL algorithm. For a value-based method with complexity-bounded function class, we show that the policy only needs to be updated for $proptooperatornamepolylog(K)$ times. In contrast to existing approaches that update the policy for at least $Omega(K)$ times, our approach drastically reduces the number of optimization calls in solving for a policy.
  arXiv  Detail & Related papers  (2021-06-14T07:36:25Z)
• Optimal Regret Algorithm for Pseudo-1d Bandit Convex Optimization [51.23789922123412]
  We study online learning with bandit feedback (i.e. learner has access to only zeroth-order oracle) where cost/reward functions admit a "pseudo-1d" structure. We show a lower bound of $min(sqrtdT, T3/4)$ for the regret of any algorithm, where $T$ is the number of rounds. We propose a new algorithm sbcalg that combines randomized online gradient descent with a kernelized exponential weights method to exploit the pseudo-1d structure effectively.
  arXiv  Detail & Related papers  (2021-02-15T08:16:51Z)
• Phase Transitions in Rate Distortion Theory and Deep Learning [5.145741425164946]
  We say that $mathcalS$ can be compressed at rate $s$ if we can achieve an error of $mathcalO(R-s)$ for encoding $mathcalS$. We show that for certain "nice" signal classes $mathcalS$, a phase transition occurs: We construct a probability measure $mathbbP$ on $mathcalS$.
  arXiv  Detail & Related papers  (2020-08-03T16:48:49Z)
• Taking a hint: How to leverage loss predictors in contextual bandits? [63.546913998407405]
  We study learning in contextual bandits with the help of loss predictors. We show that the optimal regret is $mathcalO(minsqrtT, sqrtmathcalETfrac13)$ when $mathcalE$ is known.
  arXiv  Detail & Related papers  (2020-03-04T07:36:38Z)
• Adaptive Online Learning with Varying Norms [45.11667443216861]
  We provide an online convex optimization algorithm that outputs points $w_t$ in some domain $W$. We apply this result to obtain new "full-matrix"-style regret bounds.
  arXiv  Detail & Related papers  (2020-02-10T17:22:08Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.