检索结果-内蒙古大学图书馆

A Hybrid Active-Passive Approach to Imbalanced Nonstationary Data Stream Classification

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Malialis, Kleanthis Roveri, Manuel Alippi, Cesare Panayiotou, Christos G. Polycarpou, Marios M. KIOS Research Innovation Center of Excellence Cyprus Department of Electrical and Computer Engineering University of Cyprus Nicosia Cyprus Dipartimento di Elettronica Informazione e Bioingegneria Politecnico di Milano Milan Italy Università della Svizzera Italiana Lugano Switzerland

In real-world applications, the process generating the data might suffer from nonstationary effects (e.g., due to seasonality, faults affecting sensors or actuators, and changes in the users' behaviour). These changes, often called concept drift, might induce severe (potentially catastrophic) impacts on trained learning models that become obsolete over time, and inadequate to solve the task at hand. learning in presence of concept drift aims at designing machine and deep learning models that are able to track and adapt to concept drift. Typically, techniques to handle concept drift are either active or passive, and traditionally, these have been considered to be mutually exclusive. Active techniques use an explicit drift detection mechanism, and re-train the learning algorithm when concept drift is detected. Passive techniques use an implicit method to deal with drift, and continually update the model using incremental learning. Differently from what present in the literature, we propose a hybrid alternative which merges the two approaches, hence, leveraging on their advantages. The proposed method called Hybrid-Adaptive REBAlancing (HAREBA) significantly outperforms strong baselines and state-of-the-art methods in terms of learning quality and speed;we experiment how it is effective under severe class imbalance levels too. © 2022, CC BY-NC-SA.

关键词： learning algorithms

Snn-Art: Spiking Neural Networks with Adaptive Resonance Theory

学校读者我要写书评

暂无评论

SSRN

SSRN 2022年

作者： Dagher, Issam University of Balamand Department of Computer Engineering Lebanon

In this paper we present a novel training algorithm for the Spiking Neural Networks which we call SNN-ART. Typically to train SNN, supervised learning is being used. The error-based learning is based on reducing the the difference between a target spiking train output and an actual spiking train output using variations of the backpropagation algorithm like the surrogate gradient algorithm. SNN-ART is a novel error-based spiking learning algorithm. It can have multiple hidden layers and an output *** of these layers are based on the STDP and the actual class output. The output spikes are fed into an ART (Adaptive Resonance Theory) network which compares an external output spiking train code to the internal memory of all active codes. When an output spiking train code is similar to one of the stored ART codes. It modifies that code in order to incorporate the new input into that code category. This algorithm is compared to other spiking algorithms using databases from the UCI repository. It is shown that SNN_ART provides better recognition results than other similar spiking supervised learning algorithms. © 2022, The Authors. All rights reserved.

关键词： learning algorithms

Strengthened teaching–learning-based optimization algorithm for numerical optimization tasks

学校读者我要写书评

暂无评论

Research Square

Research Square 2021年

作者： Chen, Xuefen Ye, Chunming Zhang, Yang Zhao, Lingwei Guo, Jing Ma, Kun University of Shanghai for Science and Technology China

The teaching–learning-based optimization algorithm (TLBO) is an efficient optimizer. However, it has several shortcomings such as premature convergence and stagnation at local optima. In this paper, the strengthened teaching–learning-based optimization algorithm (STLBO) is proposed to enhance the basic TLBO’s exploration and exploitation properties by introducing three strengthening mechanisms: the linear increasing teaching factor, the elite system composed of new teacher and class leader, and the Cauchy mutation. Subsequently, seven variants of STLBO are designed based on the combined deployment of the three improved mechanisms. Performance of the novel STLBOs is evaluated by implementing them on thirteen numerical optimization tasks. The results show that STLBO7 is at the top of the list, significantly better than the original TLBO. Moreover, the remaining six variants of STLBO also outperform TLBO. Finally, a set of comparisons are implemented between STLBO7 and other advanced optimization techniques. The numerical results and convergence curves prove that STLBO7 clearly outperforms other competitors, has stronger local optimal avoidance, faster convergence speed and higher solution accuracy. All the above manifests that STLBOs has improved the search performance of TLBO. Data Availability Statements: All data generated or analyzed during this study are included in this published article. © 2021, CC BY.

关键词： learning algorithms

learning to sample from censored markov random fields

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Moitra, Ankur Mossel, Elchanan Sandon, Colin Department of Mathematics Massachusetts Institute of Technology United States

We study learning Censor Markov Random Fields (abbreviated CMRFs). These are Markov Random Fields where some of the nodes are censored (not observed). We present an algorithm for learning high temperature CMRFs within o(n) transportation distance. Crucially our algorithm makes no assumption about the structure of the graph or the number or location of the observed nodes. We obtain stronger results for high girth high temperature CMRFs as well as computational lower bounds indicating that our results can not be qualitatively improved. Copyright © 2021, The Authors. All rights reserved.

关键词： learning algorithms

An experience report on machine learning reproducibility: Guidance for practitioners and tensorflow model garden contributors

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Banna, Vishnu Chinnakotla, Akhil Yan, Zhengxin Vegesana, Ani Vivek, Naveen Krishnappa, Kruthi Jiang, Wenxin Lu, Yung-Hsiang Thiruvathukal, George K. Davis, James C. Department of Electrical & Computer Engineering Purdue University Department of Computer Science Loyola University Chicago

Machine learning techniques are becoming a fundamental tool for scientific and engineering progress. These techniques are applied in contexts as diverse as astronomy and spam filtering. However, correctly applying these techniques requires careful engineering. Much attention has been paid to the technical potential;relatively little attention has been paid to the software engineering process required to bring research-based machine learning techniques into practical utility. Technology companies have supported the engineering community through machine learning frameworks such as TensorFlow and PyTorch, but the details of how to engineer complex machine learning models in these frameworks have remained hidden. To promote best practices within the engineering community, academic institutions and Google have partnered to launch a Special Interest Group on Machine learning Models (SIGMODELS) whose goal is to develop exemplary implementations of prominent machine learning models in community locations such as the TensorFlow Model Garden (TFMG). The purpose of this report is to define a process for reproducing a state-of-the-art machine learning model at a level of quality suitable for inclusion in the TFMG. We define the engineering process and elaborate on each step, from paper analysis to model release. We report on our experiences implementing the YOLO model family with a team of 26 student researchers, share the tools we developed, and describe the lessons we learned along the way. © 2021, CC BY-SA.

关键词： learning algorithms

Decentralized Personalized Federated learning: Lower Bounds and Optimal Algorithm for All Personalization Modes

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Sadiev, Abdurakhmon Borodich, Ekaterina Beznosikov, Aleksandr Dvinskikh, Darina Chezhegov, Saveliy Tappenden, Rachael Takáč, Martin Gasnikov, Alexander Moscow Institute of Physics and Technology Russia United Arab Emirates HSE University Russia Institute for Information Transmission Problems RAS Russia University of Canterbury New Zealand

This paper considers the problem of decentralized, personalized federated learning. For centralized personalized federated learning, a penalty that measures the deviation from the local model and its average, is often added to the objective function. However, in a decentralized setting this penalty is expensive in terms of communication costs, so here, a different penalty — one that is built to respect the structure of the underlying computational network — is used instead. We present lower bounds on the communication and local computation costs for this problem formulation and we also present provably optimal methods for decentralized personalized federated learning. Numerical experiments are presented to demonstrate the practical performance of our methods. Copyright © 2021, The Authors. All rights reserved.

关键词： learning algorithms

A Comprehensive Benchmark Suite for Intel SGX

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Kumar, Sandeep Panda, Abhisek Sarangi, Smruti R. School of Information Technology IIT Delhi New Delhi India Department of Computer Science and Engineering IIT Delhi New Delhi India

Trusted execution environments (TEEs) such as Intel SGX facilitate the secure execution of an application on untrusted machines. Sadly, such environments suffer from serious limitations and performance overheads in terms of writing back data to the main memory, their interaction with the OS, and the ability to issue I/O instructions. There is thus a plethora of work that focuses on improving the performance of such environments - this necessitates the need for a standard, widely accepted benchmark suite (something similar to SPEC and PARSEC). To the best of our knowledge, such a suite does not exist. Our suite, SGXGauge, contains a diverse set of workloads such as blockchain codes, secure machine learning algorithms, lightweight web servers, secure key-value stores, etc. We thoroughly characterize the behavior of the benchmark suite on a native platform and on a platform that uses a library OS-based shimming layer (GrapheneSGX). We observe that the most important metrics of interest are performance counters related to paging, memory, and TLB accesses. There is an abrupt change in performance when the memory footprint starts to exceed the size of the EPC size in Intel SGX, and the library OS does not add a significant overhead (≈ ± 10%). © 2022, CC BY.

关键词： learning algorithms

A Symmetric Loss Perspective of Reliable Machine learning

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Charoenphakdee, Nontawat Lee, Jongyeong Sugiyama, Masashi The University of Tokyo RIKEN AIP Japan RIKEN AIP The University of Tokyo Japan

When minimizing the empirical risk in binary classification, it is a common practice to replace the zero-one loss with a surrogate loss to make the learning objective feasible to optimize. Examples of well-known surrogate losses for binary classification include the logistic loss, hinge loss, and sigmoid loss. It is known that the choice of a surrogate loss can highly influence the performance of the trained classifier and therefore it should be carefully chosen. Recently, surrogate losses that satisfy a certain symmetric condition (aka., symmetric losses) have demonstrated their usefulness in learning from corrupted labels. In this article, we provide an overview of symmetric losses and their applications. First, we review how a symmetric loss can yield robust classification from corrupted labels in balanced error rate (BER) minimization and area under the receiver operating characteristic curve (AUC) maximization. Then, we demonstrate how the robust AUC maximization method can benefit natural language processing in the problem where we want to learn only from relevant keywords and unlabeled documents. Finally, we conclude this article by discussing future directions, including potential applications of symmetric losses for reliable machine learning and the design of non-symmetric losses that can benefit from the symmetric condition. Copyright © 2021, The Authors. All rights reserved.

关键词： learning algorithms

CoDiM: learning with noisy labels via contrastive semi-supervised learning

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Zhang, Xin Liu, Zixuan Xiao, Kaiwen Shen, Tian Huang, Junzhou Yang, Wei Samaras, Dimitris Han, Xiao Stony Brook University United States University of Washington United States Tencent AI Lab. China

Labels are costly and sometimes unreliable. Noisy label learning, semi-supervised learning, and contrastive learning are three different strategies for designing learning processes requiring less annotation cost. Semi-supervised learning and contrastive learning have been recently demonstrated to improve learning strategies that address datasets with noisy labels. Still, the inner connections between these fields as well as the potential to combine their strengths together have only started to emerge. In this paper, we explore further ways and advantages to fuse them. Specifically, we propose CSSL, a unified Contrastive Semi-Supervised learning algorithm, and CoDiM (Contrastive DivideMix), a novel algorithm for learning with noisy labels. CSSL leverages the power of classical semi-supervised learning and contrastive learning technologies and is further adapted to CoDiM, which learns robustly from multiple types and levels of label noise. We show that CoDiM brings consistent improvements and achieves state-of-the-art results on multiple benchmarks. © 2021, CC BY.

关键词： learning algorithms