Ideal最新文献 - Book学术

FETCH: A Memory-Efficient Replay Approach for Continual Learning in Image Classification FETCH：图像分类中持续学习的记忆高效重放方法

Ideal Pub Date : 2024-07-17 DOI: 10.1007/978-3-031-48232-8_38

Markus Weißflog, P. Protzel, Peer Neubert

引用次数: 0

ICU Mortality Prediction Using Long Short-Term Memory Networks 利用长短期记忆网络预测ICU死亡率

Ideal Pub Date : 2023-08-18 DOI: 10.1007/978-3-031-21753-1_24

Manel Mili, Asma Kerkeni, A. Abdallah, M. Hedi

引用次数: 0

Optimization of Image Acquisition for Earth Observation Satellites via Quantum Computing 通过量子计算优化地球观测卫星的图像采集

Ideal Pub Date : 2023-07-26 DOI: 10.1007/978-3-031-48232-8_1

A. Makarov, M. M. Taddei, E. Osaba, Giacomo Franceschetto, Esther Villar-Rodriguez, I. Oregi

引用次数: 0

Understanding the classes better with class-specific and rule-specific feature selection, and redundancy control in a fuzzy rule based framework 在基于模糊规则的框架中，通过特定于类和特定于规则的特征选择和冗余控制，更好地理解类

Ideal Pub Date : 2022-08-02 DOI: 10.1007/978-3-031-21753-1_33

Suchismita Das, N. Pal

引用次数: 0

Randomized K-FACs: Speeding up K-FAC with Randomized Numerical Linear Algebra 随机化K-FAC:用随机化数值线性代数加速K-FAC

Ideal Pub Date : 2022-06-30 DOI: 10.48550/arXiv.2206.15397

C. Puiu

{"title":"Randomized K-FACs: Speeding up K-FAC with Randomized Numerical Linear Algebra","authors":"C. Puiu","doi":"10.48550/arXiv.2206.15397","DOIUrl":"https://doi.org/10.48550/arXiv.2206.15397","url":null,"abstract":"K-FAC is a successful tractable implementation of Natural Gradient for Deep Learning, which nevertheless suffers from the requirement to compute the inverse of the Kronecker factors (through an eigen-decomposition). This can be very time-consuming (or even prohibitive) when these factors are large. In this paper, we theoretically show that, owing to the exponential-average construction paradigm of the Kronecker factors that is typically used, their eigen-spectrum must decay. We show numerically that in practice this decay is very rapid, leading to the idea that we could save substantial computation by only focusing on the first few eigen-modes when inverting the Kronecker-factors. Importantly, the spectrum decay happens over a constant number of modes irrespectively of the layer width. This allows us to reduce the time complexity of K-FAC from cubic to quadratic in layer width, partially closing the gap w.r.t. SENG (another practical Natural Gradient implementation for Deep learning which scales linearly in width). Randomized Numerical Linear Algebra provides us with the necessary tools to do so. Numerical results show we obtain $approx2.5times$ reduction in per-epoch time and $approx3.3times$ reduction in time to target accuracy. We compare our proposed K-FAC sped-up versions SENG, and observe that for CIFAR10 classification with VGG16_bn we perform on par with it.","PeriodicalId":113317,"journal":{"name":"Ideal","volume":"37 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-06-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123984724","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

An Implementation of the "Guess Who?" Game Using CLIP “猜猜谁?”的实现使用夹子的游戏

Ideal Pub Date : 2021-11-30 DOI: 10.1007/978-3-030-91608-4_41

Arnau Martí-Sarri, V. Rodríguez-Fernández

引用次数: 0

Stateful Optimization in Federated Learning of Neural Networks 神经网络联邦学习中的状态优化

Ideal Pub Date : 2020-11-04 DOI: 10.1007/978-3-030-62365-4_33

P. Kiss, Tomáš Horváth, Vukasin Felbab

引用次数: 0

Sequence Mining for Automatic Generation of Software Tests from GUI Event Traces 从GUI事件轨迹自动生成软件测试的序列挖掘

Ideal Pub Date : 2020-11-04 DOI: 10.1007/978-3-030-62365-4_49

Alberto Oliveira, Ricardo Freitas, A. Jorge, Vítor Amorim, Nuno Moniz, A. C. Paiva, P. Azevedo