You need to log in to edit.

You can create a new account if you don't have one.

Or, discuss a change on Slack.

You can create a new account if you don't have one.

Or, discuss a change on Slack.

no code implementations • 7 Jul 2023 • Idan Attias, Steve Hanneke, Alkis Kalavasis, Amin Karbasi, Grigoris Velegkas

Additionally, in the context of online learning we provide a dimension that characterizes the minimax instance optimal cumulative loss up to a constant factor and design an optimal online learner for realizable regression, thus resolving an open question raised by Daskalakis and Golowich in STOC '22.

no code implementations • 5 Jul 2023 • Steve Hanneke, Shay Moran, Qian Zhang

Pseudo-cubes are a structure, rooted in the work of Daniely and Shalev-Shwartz (2014), and recently shown by Brukhim, Carmon, Dinur, Moran, and Yehudayoff (2022) to characterize PAC learnability (i. e., uniform rates) for multiclass classification.

no code implementations • 22 Jun 2023 • Surbhi Goel, Steve Hanneke, Shay Moran, Abhishek Shetty

We study the problem of sequential prediction in the stochastic setting with an adversary that is allowed to inject clean-label adversarial (or out-of-distribution) examples.

no code implementations • 29 Apr 2023 • Steve Hanneke, Samory Kpotufe, Yasaman Mahdaviyeh

Theoretical studies on transfer learning or domain adaptation have so far focused on situations with a known hypothesis class or model; however in practice, some amount of model selection is usually involved, often appearing under the umbrella term of hyperparameter-tuning: for example, one may think of the problem of tuning for the right neural network architecture towards a target task, while leveraging data from a related source task.

no code implementations • 6 Apr 2023 • Maria-Florina Balcan, Steve Hanneke, Rattana Pukdee, Dravyansh Sharma

Machine learning algorithms are often used in environments which are not captured accurately even by the most carefully obtained training data, either due to the possibility of `adversarial' test-time attacks, or on account of `natural' distribution shift.

no code implementations • 30 Mar 2023 • Steve Hanneke, Shay Moran, Vinod Raman, Unique Subedi, Ambuj Tewari

We argue that the best expert has regret at most Littlestone dimension relative to the best concept in the class.

no code implementations • 27 Feb 2023 • Yuval Filmus, Steve Hanneke, Idan Mehalel, Shay Moran

We prove an analogous result for randomized learners: we show that the optimal expected mistake bound in learning a class $\mathcal{H}$ equals its randomized Littlestone dimension, which is the largest $d$ for which there exists a tree shattered by $\mathcal{H}$ whose average depth is $2d$.

no code implementations • 14 Feb 2023 • Moise Blanchard, Steve Hanneke, Patrick Jaillet

We show that optimistic universal learning for contextual bandits with adversarial rewards is impossible in general, contrary to all previously studied settings in online learning -- including standard supervised learning.

no code implementations • 12 Feb 2023 • Steve Hanneke, Aryeh Kontorovich, Guy Kornowski

We consider both the realizable and the agnostic (noisy) regression settings, proving upper and lower risk bounds in terms of the average H\"older smoothness; these rates improve upon both previously known rates even in the special case of average Lipschitz smoothness.

no code implementations • 31 Dec 2022 • Moise Blanchard, Steve Hanneke, Patrick Jaillet

Lastly, we consider the case of added continuity assumptions on rewards and show that these lead to universal consistency for significantly larger classes of data-generating processes.

no code implementations • 6 Oct 2022 • Steve Hanneke, Amin Karbasi, Mohammad Mahmoody, Idan Mehalel, Shay Moran

In this work we aim to characterize the smallest achievable error $\epsilon=\epsilon(\eta)$ by the learner in the presence of such an adversary in both realizable and agnostic settings.

no code implementations • 15 Sep 2022 • Omar Montasser, Steve Hanneke, Nathan Srebro

We present a minimax optimal learner for the problem of learning predictors robust to adversarial examples at test-time.

no code implementations • 31 Aug 2022 • Olivier Bousquet, Steve Hanneke, Shay Moran, Jonathan Shafer, Ilya Tolstikhin

We solve this problem in a principled manner, by introducing a combinatorial dimension called VCL that characterizes the best $d'$ for which $d'/n$ is a strong minimax lower bound.

no code implementations • 26 Jun 2022 • Idan Attias, Steve Hanneke

We study robustness to test-time adversarial attacks in the regression setting with $\ell_p$ losses and arbitrary perturbation sets.

no code implementations • 11 Mar 2022 • Steve Hanneke

This work provides an online learning rule that is universally consistent under processes on (X, Y) pairs, under conditions only on the X process.

no code implementations • 8 Mar 2022 • Maria-Florina Balcan, Avrim Blum, Steve Hanneke, Dravyansh Sharma

Remarkably, we provide a complete characterization of learnability in this setting, in particular, nearly-tight matching upper and lower bounds on the region that can be certified, as well as efficient algorithms for computing this region given an ERM oracle.

no code implementations • 11 Feb 2022 • Idan Attias, Steve Hanneke, Yishay Mansour

This shows that there is a significant benefit in semi-supervised robust learning even in the worst-case distribution-free model, and establishes a gap between the supervised and semi-supervised label complexities which is known not to hold in standard non-robust PAC learning.

no code implementations • 21 Jan 2022 • Moise Blanchard, Romain Cosson, Steve Hanneke

We resolve an open problem of Hanneke on the subject of universally consistent online learning with non-i. i. d.

no code implementations • 20 Oct 2021 • Omar Montasser, Steve Hanneke, Nathan Srebro

We study the problem of adversarially robust learning in the transductive setting.

no code implementations • 20 Jul 2021 • Steve Hanneke

This open problem asks whether there exists an online learning algorithm for binary classification that guarantees, for all target concepts, to make a sublinear number of mistakes, under only the assumption that the (possibly random) sequence of points X allows that such a learning algorithm can exist for that sequence.

no code implementations • 18 Jul 2021 • Noga Alon, Steve Hanneke, Ron Holzman, Shay Moran

In fact we exhibit easy-to-learn partial concept classes which provably cannot be captured by the traditional PAC theory.

no code implementations • 1 Mar 2021 • Avrim Blum, Steve Hanneke, Jian Qian, Han Shao

We study the problem of robust learning under clean-label data-poisoning attacks, where the attacker injects (an arbitrary set of) correctly-labeled examples to the training set to fool the algorithm into making mistakes on specific test instances at test time.

no code implementations • 3 Feb 2021 • Omar Montasser, Steve Hanneke, Nathan Srebro

We study the problem of learning predictors that are robust to adversarial examples with respect to an unknown perturbation set, relying instead on interaction with an adversarial attacker or access to attack oracles, examining different models for such interactions.

no code implementations • 2 Feb 2021 • Steve Hanneke, Roi Livni, Shay Moran

More precisely, given any concept class C and any hypothesis class H, we provide nearly tight bounds (up to a log factor) on the optimal mistake bounds for online learning C using predictors from H. Our bound yields an exponential improvement over the previously best known bound by Chase and Freitag (2020).

no code implementations • 9 Nov 2020 • Steve Hanneke, Aryeh Kontorovich

We analyze a family of supervised learning algorithms based on sample compression schemes that are stable, in the sense that removing points from the training set which were not selected for the compression set does not alter the resulting classifier.

no code implementations • 9 Nov 2020 • Olivier Bousquet, Steve Hanneke, Shay Moran, Ramon van Handel, Amir Yehudayoff

How quickly can a given class of concepts be learned from examples?

no code implementations • NeurIPS 2020 • Omar Montasser, Steve Hanneke, Nathan Srebro

We study the problem of reducing adversarially robust learning to standard PAC learning, i. e. the complexity of learning adversarially robust predictors using access to only a black-box non-robust learner.

no code implementations • 29 Jun 2020 • Steve Hanneke, Samory Kpotufe

A perplexing fact remains in the evolving theory on the subject: while we would hope for performance bounds that account for the contribution from multiple tasks, the vast majority of analyses result in bounds that improve at best in the number $n$ of samples per task, but most often do not improve in $N$.

no code implementations • 24 May 2020 • Olivier Bousquet, Steve Hanneke, Shay Moran, Nikita Zhivotovskiy

It has been recently shown by Hanneke (2016) that the optimal sample complexity of PAC learning for any VC class C is achieved by a particular improper learning algorithm, which outputs a specific majority-vote of hypotheses in C. This leaves the question of when this bound can be achieved by proper learning algorithms, which are restricted to always output a hypothesis from C. In this paper we aim to characterize the classes for which the optimal sample complexity can be achieved by a proper learning algorithm.

no code implementations • NeurIPS 2019 • Steve Hanneke, Samory Kpotufe

We aim to understand the value of additional labeled or unlabeled target data in transfer learning, for any given amount of source data; this is motivated by practical questions around minimizing sampling costs, whereby, target data is usually harder or costlier to acquire than source data, but can yield better accuracy.

no code implementations • 24 Jun 2019 • Steve Hanneke, Aryeh Kontorovich, Sivan Sabato, Roi Weiss

This is the first learning algorithm known to enjoy this property; by comparison, the $k$-NN classifier and its variants are not generally universally Bayes-consistent, except under additional structural assumptions, such as an inner product, a norm, finite dimension, or a Besicovitch-type property.

no code implementations • 12 Feb 2019 • Omar Montasser, Steve Hanneke, Nathan Srebro

We study the question of learning an adversarially robust predictor.

no code implementations • 3 Oct 2018 • Steve Hanneke, Aryeh Kontorovich, Menachem Sadigurschi

We show that for p in {1, infinity}, agnostic linear regression with $\ell_p$ loss admits a bounded sample compression scheme.

no code implementations • 21 May 2018 • Steve Hanneke, Aryeh Kontorovich

We establish a tight characterization of the worst-case rates for the excess risk of agnostic learning with sample compression schemes and for uniform convergence for agnostic sample compression schemes.

no code implementations • 21 May 2018 • Steve Hanneke, Aryeh Kontorovich, Menachem Sadigurschi

We give an algorithmically efficient version of the learner-to-compression scheme conversion in Moran and Yehudayoff (2016).

no code implementations • 20 Feb 2018 • Steve Hanneke, Adam Kalai, Gautam Kamath, Christos Tzamos

A generative model may generate utter nonsense when it is fit to maximize the likelihood of observed data.

no code implementations • 23 Jun 2017 • Steve Hanneke, Liu Yang

We also identify the optimal dependence on the number of pieces in the query complexity of passive testing in the special case of piecewise constant functions.

no code implementations • 5 Jun 2017 • Steve Hanneke

We are then interested in the question of whether there exist learning rules guaranteed to be universally consistent given only the assumption that universally consistent learning is possible for the given data process.

no code implementations • 29 Apr 2017 • Amit Dhurandhar, Steve Hanneke, Liu Yang

In particular, we propose an approach to provably determine the time instant from which the new/changed features start becoming relevant with respect to an output variable in an agnostic (supervised) learning setting.

no code implementations • 26 Dec 2015 • Steve Hanneke, Liu Yang

Under these conditions, we propose a learning method, and establish that for bounded VC subgraph classes, the cumulative excess risk grows sublinearly in the number of predictions, at a quantified rate.

no code implementations • 22 Dec 2015 • Steve Hanneke

This article studies the achievable guarantees on the error rates of certain learning algorithms, with particular focus on refining logarithmic factors.

no code implementations • 2 Jul 2015 • Steve Hanneke

This work establishes a new upper bound on the number of samples sufficient for PAC learning in the realizable case.

no code implementations • 20 May 2015 • Steve Hanneke, Varun Kanade, Liu Yang

Some of the results also describe an active learning variant of this setting, and provide bounds on the number of queries for the labels of points in the sequence sufficient to obtain the stated bounds on the error rates.

no code implementations • 20 May 2015 • Liu Yang, Steve Hanneke, Jaime Carbonell

We study the optimal rates of convergence for estimating a prior distribution over a VC class from a sequence of independent data sets respectively labeled by independent target functions sampled from the prior.

no code implementations • 3 Oct 2014 • Steve Hanneke, Liu Yang

This work establishes distribution-free upper and lower bounds on the minimax label complexity of active learning with general hypothesis classes, under various noise models.

no code implementations • 5 Apr 2014 • Yair Wiener, Steve Hanneke, Ran El-Yaniv

We introduce a new and improved characterization of the label complexity of disagreement-based active learning, in which the leading quantity is the version space compression set size.

no code implementations • 16 Jul 2012 • Steve Hanneke, Liu Yang

Specifically, it presents an active learning algorithm based on an arbitrary classification-calibrated surrogate loss function, along with an analysis of the number of label requests sufficient for the classifier returned by the algorithm to achieve a given risk under the 0-1 loss.

Cannot find the paper you are looking for? You can
Submit a new open access paper.

Contact us on:
hello@paperswithcode.com
.
Papers With Code is a free resource with all data licensed under CC-BY-SA.