Zobrazeno 1 - 10
of 597
pro vyhledávání: '"Larsen, Kasper"'
Autor:
Grønlund, Allan, Larsen, Kasper Green
Achieving a provable exponential quantum speedup for an important machine learning task has been a central research goal since the seminal HHL quantum algorithm for solving linear systems and the subsequent quantum recommender systems algorithm by Ke
Externí odkaz:
http://arxiv.org/abs/2411.02087
Union volume estimation is a classical algorithmic problem. Given a family of objects $O_1,\ldots,O_n \subseteq \mathbb{R}^d$, we want to approximate the volume of their union. In the special case where all objects are boxes (also known as hyperrecta
Externí odkaz:
http://arxiv.org/abs/2410.00996
Multi-distribution or collaborative learning involves learning a single predictor that works well across multiple data distributions, using samples from each during training. Recent research on multi-distribution learning, focusing on binary loss and
Externí odkaz:
http://arxiv.org/abs/2409.17567
Boosting is an extremely successful idea, allowing one to combine multiple low accuracy classifiers into a much more accurate voting classifier. In this work, we present a new and surprisingly simple Boosting algorithm that obtains a provably optimal
Externí odkaz:
http://arxiv.org/abs/2408.17148
Recent works on the parallel complexity of Boosting have established strong lower bounds on the tradeoff between the number of training rounds $p$ and the total parallel work per round $t$. These works have also presented highly non-trivial parallel
Externí odkaz:
http://arxiv.org/abs/2408.16653
PAC learning, dating back to Valiant'84 and Vapnik and Chervonenkis'64,'74, is a classic model for studying supervised learning. In the agnostic setting, we have access to a hypothesis set $\mathcal{H}$ and a training set of labeled samples $(x_1,y_1
Externí odkaz:
http://arxiv.org/abs/2407.19777
Developing an optimal PAC learning algorithm in the realizable setting, where empirical risk minimization (ERM) is suboptimal, was a major open problem in learning theory for decades. The problem was finally resolved by Hanneke a few years ago. Unfor
Externí odkaz:
http://arxiv.org/abs/2403.08831
We provide efficient replicable algorithms for the problem of learning large-margin halfspaces. Our results improve upon the algorithms provided by Impagliazzo, Lei, Pitassi, and Sorrell [STOC, 2022]. We design the first dimension-independent replica
Externí odkaz:
http://arxiv.org/abs/2402.13857
For constants $\gamma \in (0,1)$ and $A\in (1,\infty)$, we prove existence and uniqueness of a solution to the singular and path-dependent Riccati-type ODE \begin{align*} \begin{cases} h'(y) = \frac{1+\gamma}{y}\big( \gamma - h(y)\big)+h(y)\frac{\gam
Externí odkaz:
http://arxiv.org/abs/2402.07185
In boosting, we aim to leverage multiple weak learners to produce a strong learner. At the center of this paradigm lies the concept of building the strong learner as a voting classifier, which outputs a weighted majority vote of the weak learners. Wh
Externí odkaz:
http://arxiv.org/abs/2402.02976