<div dir="ltr"><p class="MsoNormal" style="margin:0cm;font-size:12pt;font-family:Aptos,sans-serif"><span style="font-family:Calibri,sans-serif">Dear AI scientist/engineer/student/enthusiast,</span></p><p class="MsoNormal" style="margin:0cm;font-size:12pt;font-family:Aptos,sans-serif"><span style="font-family:Calibri,sans-serif"> </span></p><p class="MsoNormal" style="margin:0cm;font-size:12pt;font-family:Aptos,sans-serif"><b><span style="font-family:Calibri,sans-serif">Professor Ioannis Pitas</span></b><span style="font-family:Calibri,sans-serif"> (IEEE Fellow and Distinguished Lecturer, Chair of <a href="https://www.i-aida.org/" target="_blank" style="color:rgb(70,120,134)">AIDA</a>, EURASIP fellow and Director of <a href="https://aiia.csd.auth.gr/" target="_blank" style="color:rgb(70,120,134)">AIIA</a> Lab, Aristotle University of Thessaloniki, Greece) will deliver the e – lecture: “<b>Decentralized DNN architectures for big data analysis</b>” on <b>22nd October 2024, 17:00 CEST.</b></span></p><p class="MsoNormal" style="margin:0cm;font-size:12pt;font-family:Aptos,sans-serif"><span style="font-family:Calibri,sans-serif"> </span></p><p class="MsoNormal" style="margin:0cm;font-size:12pt;font-family:Aptos,sans-serif"><span style="font-family:Calibri,sans-serif">This lecture overviews decentralized and distributed DNN architectures for DNN training and inference. Big data analysis can be greatly facilitated, if decentralized/distributed DNN agent deployment is employed.</span></p><p class="MsoNormal" style="margin:0cm;font-size:12pt;font-family:Aptos,sans-serif"><span style="font-family:Calibri,sans-serif">Such DNN agents can interact with each other for their training and/or inference following the human Teacher-Student Learning-by-Education paradigm to achieve superior performance, e.g.,</span></p><p class="MsoNormal" style="margin:0cm;font-size:12pt;font-family:Aptos,sans-serif"><span style="font-family:Calibri,sans-serif">in continual learning, knowledge distillation or federated learning settings. Significant speedup, privacy protection, robustness and classification performance increase can be obtained.</span></p><p class="MsoNormal" style="margin:0cm;font-size:12pt;font-family:Aptos,sans-serif"><span style="font-family:Calibri,sans-serif">·  Join via Zoom: Zoom <a href="https://authgr.zoom.us/j/91937490266?pwd=IJaviveYw7aaO1unNtb3cbNaQl6xMU.1" title="https://authgr.zoom.us/j/91937490266?pwd=IJaviveYw7aaO1unNtb3cbNaQl6xMU.1" target="_blank" style="color:rgb(70,120,134)">link</a> & passcode: 114050</span></p><p class="MsoNormal" style="margin:0cm;font-size:12pt;font-family:Aptos,sans-serif"><span style="font-family:Calibri,sans-serif">·  For more details, click <a href="https://www.i-aida.org/events/decentralized-dnn-architectures-for-big-data-analysis/" title="https://www.i-aida.org/events/decentralized-dnn-architectures-for-big-data-analysis/" target="_blank" style="color:rgb(70,120,134)">here</a>.</span></p><p class="MsoNormal" style="margin:0cm;font-size:12pt;font-family:Aptos,sans-serif"><span style="font-family:Calibri,sans-serif"> </span></p><p class="MsoNormal" style="margin:0cm;font-size:12pt;font-family:Aptos,sans-serif"><span style="font-family:Calibri,sans-serif">The <b><u><a href="http://www.i-aida.org/" target="_blank" style="color:rgb(70,120,134)">International AI Doctoral Academy (AIDA)</a></u></b>, a joint initiative of the European R&D projects <u><a href="https://ai4media.eu/" target="_blank" style="color:rgb(70,120,134)">AI4Media</a></u>, <u><a href="https://www.elise-ai.eu/" target="_blank" style="color:rgb(70,120,134)">ELISE</a></u>, <u><a href="https://www.humane-ai.eu/" target="_blank" style="color:rgb(70,120,134)">Humane AI Net</a></u>, <u><a href="https://tailor-network.eu/" target="_blank" style="color:rgb(70,120,134)">TAILOR</a>, <a href="https://www.vision4ai.eu/" target="_blank" style="color:rgb(70,120,134)">VISION</a></u>, is very pleased to offer you top quality scientific lectures </span></p><p class="MsoNormal" style="margin:0cm;font-size:12pt;font-family:Aptos,sans-serif"><span style="font-family:Calibri,sans-serif">in the framework of <u><a href="https://www.i-aida.org/event_cat/ai-lectures/" target="_blank" style="color:rgb(70,120,134)">AIDA AI Excellence Lecture Series</a></u> on several current hot AI topics. Lectures will be offered alternatingly by:</span></p><ul type="disc" style="margin-top:0cm;margin-bottom:0cm"><li class="MsoNormal" style="margin:0cm;font-size:12pt;font-family:Aptos,sans-serif"><span style="font-family:Calibri,sans-serif">Top highly cited senior AI scientists internationally or</span></li><li class="MsoNormal" style="margin:0cm;font-size:12pt;font-family:Aptos,sans-serif"><span style="font-family:Calibri,sans-serif">Young AI scientists with promise of excellence (AI sprint lectures)</span></li></ul><p class="MsoNormal" style="margin:0cm;font-size:12pt;font-family:Aptos,sans-serif"><span style="font-family:Calibri,sans-serif">These lectures are disseminated through multiple channels and email lists (we apologize if you received it through various channels). If you want to stay informed on future lectures, you can register in the email lists <u><a href="https://lists.auth.gr/sympa/info/aida" target="_blank" style="color:rgb(70,120,134)">AIDA email list</a></u> and <u><a href="https://lists.auth.gr/sympa/info/cvml" target="_blank" style="color:rgb(70,120,134)">CVML email list</a></u>.</span></p><p class="MsoNormal" style="margin:0cm;font-size:12pt;font-family:Aptos,sans-serif"><span style="font-family:Calibri,sans-serif"> </span></p><p style="margin:0cm;text-align:justify"></p><p class="MsoNormal" style="margin:0cm;font-size:12pt;font-family:Aptos,sans-serif"><span style="font-family:Calibri,sans-serif">Best regards<br>Profs. N. Sebe, M. Chetouani, P. Flach, B. O’Sullivan, I. Pitas, J. Stefanowski<br>AIDA AI Excellence Lecture Series committee members</span></p></div>