博舍

人工智能研究院学术报告 第2023 烟台大学人工智能研究院

人工智能研究院学术报告 第2023

   报告题目  

InnovatingandInterpretingNeuralNetworks

   报告摘要  

Deeplearninghasrecentlyachievedhugesuccessinmanyapplications,includingnaturallanguageprocessing,computervisionandmore.Inthesecases,deeplearningcanoutperformorcompetewithhumans.Itiswidelyrecognizedthatmachinelearning,especiallydeeplearning,isaparadigmshiftinmanyfields.However,therearestillmanychallengesahead.Ononehand,overthepastyears,majoreffortshavebeendedicatedtoarchitectureinnovationsinthefieldofneuralnetworks,leadingtomanyadvancedmodels.Althoughdeeplearningisinspiredbythecomputationoftheneuralsystem,currentdeeplearningsystemsfallshortofreflectingneuraldiversity.Ontheotherhand,despitethefactthatdeeplearningperformsquitewellinpractice,itisdifficulttoexplainitsunderlyingmechanismandunderstanditsbehaviors.Thesuccessofdeeplearningisnotwellunderpinnedbytheeffectivetheory.Lackinginterpretabilityhasbecomeaprimaryobstacletothewidespreadtranslationandfurtherdevelopmentofdeeplearningtechniques.Inthisproject,weproposequadraticneuronstoaddresstheneuraldiversityproblemindeeplearning,whereinnerproducts(whicharelinearoperations)arereplacedwithquadraticcounterpartswhosenon-linearityenhancestheexpressiveabilityoftheneuron.Further,weproposesoftthresholdingtoreplaceReLUactivationforsignalprocessingtasks.Wewillevaluatetheirfeasibilityinpracticalcomputervisionproblemsaswellasmedicalimagingproblems,therebyenrichingmachinelearningarmory.Wewillalsodevelopinterpretationmethodsfortheinnerworkingofneuralnetworksandaccountabletheoriesforthesuccessofdeepnetworks.

版权声明:本文内容由互联网用户自发贡献,该文观点仅代表作者本人。本站仅提供信息存储空间服务,不拥有所有权,不承担相关法律责任。如发现本站有涉嫌抄袭侵权/违法违规的内容,一经查实,本站将立刻删除。

上一篇

下一篇