التفاصيل البيبلوغرافية
العنوان: |
Learning with incremental iterative regularization |
المؤلفون: |
Lorenzo Rosasco, Silvia Villa |
المساهمون: |
edited by C. Cortes, N.D. Lawrence, D.D. Lee, M. Sugiyama, R. Garnett, Rosasco, Lorenzo, Villa, Silvia |
بيانات النشر: |
Neural Information Processing Systems Foundation |
سنة النشر: |
2015 |
المجموعة: |
Università degli Studi di Genova: CINECA IRIS |
مصطلحات موضوعية: |
Algorithms, Artificial intelligence, Gradient methods, Information science, Learning systems, Least squares approximations, Sampling, Stochastic systems, Almost sure convergence, Finite samples. Iterative regularization, Least Square, Regularization parameters, Statistical learning, Stochastic gradient techniques, Universal consistency, Iterative methods |
الوصف: |
Within a statistical learning setting, we propose and study an iterative regularization algorithm for least squares defined by an incremental gradient method. In particular, we show that, if all other parameters are fixed a priori, the number of passes over the data (epochs) acts as a regularization parameter, and prove strong universal consistency, i.e. almost sure convergence of the risk, as well as sharp finite sample bounds for the iterates. Our results are a step towards understanding the effect of multiple epochs in stochastic gradient techniques in machine learning and rely on integrating statistical and optimization results |
نوع الوثيقة: |
conference object |
وصف الملف: |
ELETTRONICO |
اللغة: |
English |
العلاقة: |
info:eu-repo/semantics/altIdentifier/wos/WOS:000450913103053; ispartofbook:Advances in Neural Information Processing Systems 28 (NIPS 2015); 29th Annual Conference on Neural Information Processing Systems - NIPS 2015; firstpage:1630; lastpage:1638; numberofpages:9; serie:ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS; http://hdl.handle.net/11567/888561Test; info:eu-repo/semantics/altIdentifier/scopus/2-s2.0-84965154543 |
الإتاحة: |
http://hdl.handle.net/11567/888561Test |
حقوق: |
info:eu-repo/semantics/openAccess |
رقم الانضمام: |
edsbas.46FAC71D |
قاعدة البيانات: |
BASE |