The University of Macau (UM) held a talk titled ‘A Preliminary Exploration to Learnware’ today (25 March) as part of the University Lecture Series. The speaker was Zhou Zhi-Hua, vice president of Nanjing University and president of the International Joint Conferences on Artificial Intelligence Organization (IJCAI) Trustee, who discussed the innovative concept of ‘learnware’ in machine learning.
Speaking at the event, UM Rector Yonghua Song highlighted Prof Zhou’s significant contributions to artificial intelligence (AI), machine learning, and data mining, particularly in the areas of ensemble learning, multi-label learning, and weakly supervised learning. Prof Zhou’s research achievements have earned him international recognition, including fellowships in a number of prestigious societies, as well as national and international awards.
During the talk, Prof Zhou explained the new concept of ‘learnware’ in machine learning. He mentioned the idea of ‘Learnware = Model + Specification’ and discussed how small models can be reused to achieve or even exceed the original objectives. Prof Zhou also engaged with the audience during the Q&A and discussion session moderated by Xu Cheng-Zhong, dean of the Faculty of Science and Technology and interim director of the Institute of Collaborative Innovation of UM.
Source: Rector’s Office
澳門大學今（25）日舉行“大學講壇”，邀請南京大學副校長、國際人工智能聯合會理事會主席周志華以“學件的初步探索”為題，圍繞“學件”這一機器學習新概念發表演說。
澳大校長宋永華表示，周志華是世界頂尖的人工智能學者，他在人工智能、機器學習、數據挖掘領域的研究貢獻良多，特別在集成學習、多標簽學習、弱監督學習等前沿研究方向取得了突破性成果。周志華的研究成就亦獲得國際廣泛認可，包括當選為ACM、AAAI、AAAS、IEEE等學會會士及歐洲科學院外籍院士，並獲中國國家自然科學獎、IEEE計算機學會Edward J. McCluskey技術成就獎等。
講座上，周志華為與會者深入淺出地剖析“學件”作為機器學習的新概念。講座圍繞“學件=模型+規約”的新概念，探討如何通過複用小型模型完成甚至超越原始設定的任務目標。他還重點介紹了“規約”這一關鍵元素的重要性——令訓練好的模型能夠在未來根據用戶的需求被準確識別並加以複用。周志華的演講內容充滿前瞻性和啟發性，不僅為與會者提供了豐富的知識和啟發，同時激發了他們對人工智能技術的深層次理解與思考。
問答與交流環節由澳大科技學院院長、協同創新研究院代院長須成忠主持。周志華和與會者交流互動，討論氣氛熱烈。
新聞來源：校長辦公室
Zhou Zhi-Hua
周志華
Yonghua Song
宋永華
The Q&A and discussion session
問答與交流環節