On Agnostic PAC Learning in the Small Error Regime

• URL: http://arxiv.org/abs/2502.09496v1
• Date: Thu, 13 Feb 2025 17:03:03 GMT
• Title: On Agnostic PAC Learning in the Small Error Regime
• Authors: Julian Asilis, Mikael Møller Høgsgaard, Grigoris Velegkas,
• Abstract summary: Empirical Risk Minimization learners are suboptimal in the realizable case yet optimal in the agnostic case. Work of Hanneke, Larsen, and Zhivotovskiy addresses this shortcoming by including $tau$ as a parameter in the error term. We show that our learner achieves tightness of the error lower bound $tau + Omega left(sqrtfractau))m + fracd + log (1 / delta)m right)$ for a constant $c leq 2.1$
• Score: 4.422219522591412
• License:
• Abstract: Binary classification in the classic PAC model exhibits a curious phenomenon: Empirical Risk Minimization (ERM) learners are suboptimal in the realizable case yet optimal in the agnostic case. Roughly speaking, this owes itself to the fact that non-realizable distributions $\mathcal{D}$ are simply more difficult to learn than realizable distributions -- even when one discounts a learner's error by $\mathrm{err}(h^*_{\mathcal{D}})$, the error of the best hypothesis in $\mathcal{H}$ for $\mathcal{D}$. Thus, optimal agnostic learners are permitted to incur excess error on (easier-to-learn) distributions $\mathcal{D}$ for which $\tau = \mathrm{err}(h^*_{\mathcal{D}})$ is small. Recent work of Hanneke, Larsen, and Zhivotovskiy (FOCS `24) addresses this shortcoming by including $\tau$ itself as a parameter in the agnostic error term. In this more fine-grained model, they demonstrate tightness of the error lower bound $\tau + \Omega \left(\sqrt{\frac{\tau (d + \log(1 / \delta))}{m}} + \frac{d + \log(1 / \delta)}{m} \right)$ in a regime where $\tau > d/m$, and leave open the question of whether there may be a higher lower bound when $\tau \approx d/m$, with $d$ denoting $\mathrm{VC}(\mathcal{H})$. In this work, we resolve this question by exhibiting a learner which achieves error $c \cdot \tau + O \left(\sqrt{\frac{\tau (d + \log(1 / \delta))}{m}} + \frac{d + \log(1 / \delta)}{m} \right)$ for a constant $c \leq 2.1$, thus matching the lower bound when $\tau \approx d/m$. Further, our learner is computationally efficient and is based upon careful aggregations of ERM classifiers, making progress on two other questions of Hanneke, Larsen, and Zhivotovskiy (FOCS `24). We leave open the interesting question of whether our approach can be refined to lower the constant from 2.1 to 1, which would completely settle the complexity of agnostic learning.

Related papers

• Revisiting Agnostic PAC Learning [30.67561230812141]
  PAC learning, dating back to Valiant'84 and Vapnik and Chervonenkis'64,'74, is a classic model for studying supervised learning. Empirical Risk Minimization (ERM) is a natural learning algorithm, where one simply outputs the hypothesis from $mathcalH$ making the fewest mistakes on the training data. We revisit PAC learning and first show that ERM is in fact sub-optimal if we treat the performance of the best hypothesis, denoted $tau:=Pr_mathcalD[hstar_math
  arXiv  Detail & Related papers  (2024-07-29T08:20:49Z)
• Cryptographic Hardness of Learning Halfspaces with Massart Noise [59.8587499110224]
  We study the complexity of PAC learning halfspaces in the presence of Massart noise. We show that no-time Massart halfspace learners can achieve error better than $Omega(eta)$, even if the optimal 0-1 error is small.
  arXiv  Detail & Related papers  (2022-07-28T17:50:53Z)
• Learning a Single Neuron with Adversarial Label Noise via Gradient Descent [50.659479930171585]
  We study a function of the form $mathbfxmapstosigma(mathbfwcdotmathbfx)$ for monotone activations. The goal of the learner is to output a hypothesis vector $mathbfw$ that $F(mathbbw)=C, epsilon$ with high probability.
  arXiv  Detail & Related papers  (2022-06-17T17:55:43Z)
• Threshold Phenomena in Learning Halfspaces with Massart Noise [56.01192577666607]
  We study the problem of PAC learning halfspaces on $mathbbRd$ with Massart noise under Gaussian marginals. Our results qualitatively characterize the complexity of learning halfspaces in the Massart model.
  arXiv  Detail & Related papers  (2021-08-19T16:16:48Z)
• Hardness of Learning Halfspaces with Massart Noise [56.98280399449707]
  We study the complexity of PAC learning halfspaces in the presence of Massart (bounded) noise. We show that there an exponential gap between the information-theoretically optimal error and the best error that can be achieved by a SQ algorithm.
  arXiv  Detail & Related papers  (2020-12-17T16:43:11Z)
• An Algorithm for Learning Smaller Representations of Models With Scarce Data [0.0]
  We present a greedy algorithm for solving binary classification problems in situations where the dataset is too small or not fully representative. It relies on a trained model with loose accuracy constraints, an iterative hyperparameter pruning procedure, and a function used to generate new data.
  arXiv  Detail & Related papers  (2020-10-15T19:17:51Z)
• Near-Optimal SQ Lower Bounds for Agnostically Learning Halfspaces and ReLUs under Gaussian Marginals [49.60752558064027]
  We study the fundamental problems of agnostically learning halfspaces and ReLUs under Gaussian marginals. Our lower bounds provide strong evidence that current upper bounds for these tasks are essentially best possible.
  arXiv  Detail & Related papers  (2020-06-29T17:10:10Z)
• Learning and Testing Variable Partitions [13.575794982844222]
  We show that $mathcalO(k n2)(delta + epsilon)$ can be learned in time $tildemathcalO(n2 mathrmpoly (1/epsilon)$ for any $epsilon > 0$. We also show that even two-sided testers require $Omega(n)$ queries when $k = 2$.
  arXiv  Detail & Related papers  (2020-03-29T10:12:32Z)
• Agnostic Q-learning with Function Approximation in Deterministic Systems: Tight Bounds on Approximation Error and Sample Complexity [94.37110094442136]
  We study the problem of agnostic $Q$-learning with function approximation in deterministic systems. We show that if $delta = Oleft(rho/sqrtdim_Eright)$, then one can find the optimal policy using $Oleft(dim_Eright)$.
  arXiv  Detail & Related papers  (2020-02-17T18:41:49Z)

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.