A generalized online algorithm for translation and scale invariant prediction with expert advice

K Gokcesu, H Gokcesu - arxiv preprint arxiv:2009.04372, 2020 - arxiv.org
In this work, we aim to create a completely online algorithmic framework for prediction with
expert advice that is translation-free and scale-free of the expert losses. Our goal is to create …

Recursive experts: An efficient optimal mixture of learning systems in dynamic environments

K Gokcesu, H Gokcesu - arxiv preprint arxiv:2009.09249, 2020 - arxiv.org
Sequential learning systems are used in a wide variety of problems from decision making to
optimization, where they provide a'belief'(opinion) to nature, and then update this belief …

Low regret binary sampling method for efficient global optimization of univariate functions

K Gokcesu, H Gokcesu - arxiv preprint arxiv:2201.07164, 2022 - arxiv.org
In this work, we propose a computationally efficient algorithm for the problem of global
optimization in univariate loss functions. For the performance evaluation, we study the …

Nonconvex extension of generalized huber loss for robust learning and pseudo-mode statistics

K Gokcesu, H Gokcesu - arxiv preprint arxiv:2202.11141, 2022 - arxiv.org
We propose an extended generalization of the pseudo Huber loss formulation. We show that
using the log-exp transform together with the logistic function, we can create a loss which …

Optimal and efficient algorithms for general mixable losses against switching oracles

K Gokcesu, H Gokcesu - arxiv preprint arxiv:2108.06411, 2021 - arxiv.org
We investigate the problem of online learning, which has gained significant attention in
recent years due to its applicability in a wide range of fields from machine learning to game …

Efficient Minimax Optimal Global Optimization of Lipschitz Continuous Multivariate Functions

K Gokcesu, H Gokcesu - arxiv preprint arxiv:2206.02383, 2022 - arxiv.org
In this work, we propose an efficient minimax optimal global optimization algorithm for
multivariate Lipschitz continuous functions. To evaluate the performance of our approach …

Efficient, anytime algorithms for calibration with isotonic regression under strictly convex losses

K Gokcesu, H Gokcesu - arxiv preprint arxiv:2111.00468, 2021 - arxiv.org
We investigate the calibration of estimations to increase performance with an optimal
monotone transform on the estimator outputs. We start by studying the traditional square …

Natural Hierarchical Cluster Analysis by Nearest Neighbors with Near-Linear Time Complexity

K Gokcesu, H Gokcesu - arxiv preprint arxiv:2203.08027, 2022 - arxiv.org
We propose a nearest neighbor based clustering algorithm that results in a naturally defined
hierarchy of clusters. In contrast to the agglomerative and divisive hierarchical clustering …