Online linear regression in dynamic environments via discounting
A Jacobsen, A Cutkosky - arXiv preprint arXiv:2405.19175, 2024 - arxiv.org
We develop algorithms for online linear regression which achieve optimal static and
dynamic regret guarantees\emph {even in the complete absence of prior knowledge}. We …
dynamic regret guarantees\emph {even in the complete absence of prior knowledge}. We …
Scale-free unconstrained online learning for curved losses
A sequence of works in unconstrained online convex optimisation have investigated the
possibility of adapting simultaneously to the norm U of the comparator and the maximum …
possibility of adapting simultaneously to the norm U of the comparator and the maximum …
Stochastic online linear regression: the forward algorithm to replace ridge
R Ouhamma, OA Maillard… - Advances in Neural …, 2021 - proceedings.neurips.cc
We consider the problem of online linear regression in the stochastic setting. We derive high
probability regret bounds for online $\textit {ridge} $ regression and the $\textit {forward} …
probability regret bounds for online $\textit {ridge} $ regression and the $\textit {forward} …
Online forgetting process for linear regression models
Motivated by the EU's" Right To Be Forgotten" regulation, we initiate a study of statistical
data deletion problems where users' data are accessible only for a limited period of time …
data deletion problems where users' data are accessible only for a limited period of time …
The Gain from Ordering in Online Learning
We study fixed-design online learning where the learner is allowed to choose the order of
the datapoints in order to minimize their regret (aka self-directed online learning). We focus …
the datapoints in order to minimize their regret (aka self-directed online learning). We focus …
Online instrumental variable regression: Regret analysis and bandit feedback
R Della Vecchia, D Basu - arXiv preprint arXiv:2302.09357, 2023 - arxiv.org
Endogeneity, ie the dependence between noise and covariates, is a common phenomenon
in real data due to omitted variables, strategic behaviours, measurement errors etc. In …
in real data due to omitted variables, strategic behaviours, measurement errors etc. In …
Quasi-newton steps for efficient online exp-concave optimization
Z Mhammedi, K Gatmiry - The Thirty Sixth Annual …, 2023 - proceedings.mlr.press
The aim of this paper is to design computationally-efficient and optimal algorithms for the
online and stochastic exp-concave optimization settings. Typical algorithms for these …
online and stochastic exp-concave optimization settings. Typical algorithms for these …
Online hierarchical forecasting for power consumption data
M Brégère, M Huard - International Journal of Forecasting, 2022 - Elsevier
This paper proposes a three-step approach to forecasting time series of electricity
consumption at different levels of household aggregation. These series are linked by …
consumption at different levels of household aggregation. These series are linked by …
Locality Sensitive Sparse Encoding for Learning World Models Online
Acquiring an accurate world model online for model-based reinforcement learning (MBRL)
is challenging due to data nonstationarity, which typically causes catastrophic forgetting for …
is challenging due to data nonstationarity, which typically causes catastrophic forgetting for …
Efficient online learning with kernels for adversarial large scale problems
We are interested in a framework of online learning with kernels for low-dimensional, but
large-scale and potentially adversarial datasets. We study the computational and theoretical …
large-scale and potentially adversarial datasets. We study the computational and theoretical …