Related papers: Online Learning for Non-Stationary A/B Tests
We provide a new efficient adaptive algorithm for performing phase estimation that does not require that the user infer the bits of the eigenphase in reverse order; rather it directly infers the phase and estimates the uncertainty in the…
Many modern datasets don't fit neatly into $n \times p$ matrices, but most techniques for measuring statistical stability expect rectangular data. We study methods for stability assessment on non-rectangular data, using statistical learning…
eBay's experimentation platform runs hundreds of A/B tests on any given day. The platform integrates with the tracking infrastructure and customer experience servers, provides the sampling service for experiments, and has the responsibility…
Big data analytics is gaining massive momentum in the last few years. Applying machine learning models to big data has become an implicit requirement or an expectation for most analysis tasks, especially on high-stakes applications.Typical…
We study supervisory switching control for partially-observed linear dynamical systems. The objective is to identify and deploy the best controller for the unknown system by periodically selecting among a collection of $N$ candidate…
Dynamic Binary Instrumentation (DBI) is the set of techniques that enable instrumentation of programs at run-time, making it possible to monitor and modify the execution of compiled binaries or entire systems. DBI is used for countless…
Instruction Fine-Tuning enhances pre-trained language models from basic next-word prediction to complex instruction-following. However, existing One-off Instruction Fine-Tuning (One-off IFT) method, applied on a diverse instruction, may not…
Guided exploration with expert demonstrations improves data efficiency for reinforcement learning, but current algorithms often overuse expert information. We propose a novel algorithm to speed up Q-learning with the help of a limited…
While the identification of nonlinear dynamical systems is a fundamental building block of model-based reinforcement learning and feedback control, its sample complexity is only understood for systems that either have discrete states and…
Interactive applications incorporating high-data rate sensing and computer vision are becoming possible due to novel runtime systems and the use of parallel computation resources. To allow interactive use, such applications require careful…
Most generalization bounds in learning theory are based on some measure of the complexity of the hypothesis class used, independently of any algorithm. In contrast, the notion of algorithmic stability can be used to derive tight…
A/B testing, or controlled experiments, is the gold standard approach to causally compare the performance of algorithms on online platforms. However, conventional Bernoulli randomization in A/B testing faces many challenges such as…
Emphatic algorithms are temporal-difference learning algorithms that change their effective state distribution by selectively emphasizing and de-emphasizing their updates on different time steps. Recent works by Sutton, Mahmood and White…
This paper deals with the problem of finite-time learning for unknown discrete-time nonlinear systems' dynamics, without the requirement of the persistence of excitation. Two finite-time concurrent learning methods are presented to…
Innovations across science and industry are evaluated using randomized trials (a.k.a. A/B tests). While simple and robust, such static designs are inefficient or infeasible for testing many hypotheses. Adaptive designs can greatly improve…
Online controlled experiments are a crucial tool to allow for confident decision-making in technology companies. A North Star metric is defined (such as long-term revenue or user retention), and system variants that statistically…
Continual learning is an emerging paradigm in machine learning, wherein a model is exposed in an online fashion to data from multiple different distributions (i.e. environments), and is expected to adapt to the distribution change.…
Classification algorithms aim to predict an unknown label (e.g., a quality class) for a new instance (e.g., a product). Therefore, training samples (instances and labels) are used to deduct classification hypotheses. Often, it is relatively…
Reinforcement Learning (RL) and continuous nonlinear control have been successfully deployed in multiple domains of complicated sequential decision-making tasks. However, given the exploration nature of the learning process and the presence…
The training of autonomous agents often requires expensive and unsafe trial-and-error interactions with the environment. Nowadays several data sets containing recorded experiences of intelligent agents performing various tasks, spanning…