paper-with-me

홈 › Papers

SpotTune: Leveraging Transient Resources for Cost-efficient Hyper-parameter Tuning in the Public Cloud

2020-12-07 · Yan Li, Bo An, Junming Ma, Donggang Cao, Yasha Wang, Hong Mei

Hyper-parameter tuning (HPT) is crucial for many machine learning (ML) algorithms. But due to the large searching space, HPT is usually time-consuming and resource-intensive. Nowadays, many researchers use public cloud resources to train machine learning models, convenient yet expensive. How to speed up the HPT process while at the same time reduce cost is very important for cloud ML users. In this paper, we propose SpotTune, an approach that exploits transient revocable resources in the public cloud with some tailored strategies to do HPT in a parallel and cost-efficient manner. Orchestrating the HPT process upon transient servers, SpotTune uses two main techniques, fine-grained cost-aware resource provisioning, and ML training trend predicting, to reduce the monetary cost and runtime of HPT processes. Our evaluations show that SpotTune can reduce the cost by up to 90% and achieve a 16.61x performance-cost rate improvement.

📄 PDF Abstract BibTeX arXiv:2012.03576

Code (0)

등록된 구현이 없습니다.

Tasks

BIG-bench Machine Learning

Similar Papers 제목 키워드 기반

SpotTune: Transfer Learning through Adaptive Fine-tuning

2018-11-21 · CVPR 2019 6 · Yunhui Guo, Honghui Shi, Abhishek Kumar, Kristen Grauman 외

Transfer learning, which allows a source task to affect the inductive bias of the target task, is widely used in computer vision. The typical way of conducting transfer learning with deep neural networks is to fine-tune …

Inductive BiasTransfer Learning

Anatomy of Domain Shift Impact on U-Net Layers in MRI Segmentation

2021-07-10 · Ivan Zakazov, Boris Shirokikh, Alexey Chernyavskiy, Mikhail Belyaev

Domain Adaptation (DA) methods are widely used in medical image segmentation tasks to tackle the problem of differently distributed train (source) and test (target) data. We consider the supervised DA task with a limited…

AnatomyDomain AdaptationDomain GeneralizationImage Segmentation+3

Simple and Scalable Parallelized Bayesian Optimization

2020-06-24 · Masahiro Nomura

In recent years, leveraging parallel and distributed computational resources has become essential to solve problems of high computational cost. Bayesian optimization (BO) has shown attractive results in those expensive-t…

Bayesian OptimizationBIG-bench Machine LearningHyperparameter Optimization

Ca2+ release via IP3 receptors shapes the cytosolic Ca2+ transient for hypertrophic signalling in ventricular cardiomyocytes

2019-02-13 · Hilary Hunt, Agne Tilunaite, Greg Bass, Christian Soeller 외

Calcium (Ca2+) plays a central role in mediating both contractile function and hypertrophic signalling in ventricular cardiomyocytes. L-type Ca2+ channels trigger release of Ca2+ from ryanodine receptors (RyRs) for cellu…

Lightweight Deep Learning Architecture for MPI Correction and Transient Reconstruction

2021-11-29 · Adriano Simonetto, Gianluca Agresti, Pietro Zanuttigh, Henrik Schäfer

Indirect Time-of-Flight cameras (iToF) are low-cost devices that provide depth images at an interactive frame rate. However, they are affected by different error sources, with the spotlight taken by Multi-Path Interferen…

Deep Learning