Invited Talk: Recent Advances on Neural Architecture Search
Yanan Sun
Sichuan University, China
Technically co-sponsored by the IEEE SP Chengdu chapter, we are excited to invite Prof. Yanan Sun from Sichuan University, China, to present an academic talk titled "Recent Advances on Neural Architecture Search." This event is designed for students and faculty from major universities in Chengdu, including the University of Electronic Science and Technology of China, Sichuan University, and Southwest Jiaotong University, as well as IEEE members in the region.
In his talk, Prof. Sun will delve into the latest developments in neural architecture search (NAS), a transformative area within artificial intelligence and machine learning that automates the design of neural networks. He will discuss innovative methodologies and algorithms that have emerged, showcasing how these advancements can optimize model performance and efficiency.
Participants can expect a comprehensive overview of the state-of-the-art techniques in NAS, including practical applications and case studies that highlight the impact of these technologies across various domains. The session aims to foster an engaging environment for academic exchange, encouraging collaboration and discussion among attendees.
We look forward to welcoming a diverse audience eager to deepen their understanding of this cutting-edge topic and explore potential research opportunities in neural architecture search. Join us for what promises to be an enlightening and inspiring presentation!
Date and Time
Location
Hosts
Registration
- Date: 25 May 2024
- Time: 03:00 AM UTC to 04:00 AM UTC
-
Add Event to Calendar
- Chengdu Taihe International Hotel
- BINJIANG ZHONG ROAD
- Chengdu, Sichuan
- China 610054
- Building: Chengdu Taihe International Hotel
- Room Number: Room: Versailles
- Contact Event Host
- Co-sponsored by University of Electronic Science and Technology of China
- Starts 30 April 2024 04:00 PM UTC
- Ends 19 May 2024 04:00 PM UTC
- No Admission Charge
Speakers
Yanan SUN of Sichuan University
Recent Advances on Neural Architecture Search
Large language models have become a competitive focus in the global field of artificial intelligence. However, in specific scenarios, it is often necessary to redesign models to meet usage requirements. Neural Architecture Search (NAS) can automatically design neural networks for different tasks. This speech outlines the three components of NAS: the search space, the search strategy, and the performance evaluation strategy. It also points out the shortcomings of existing performance evaluation strategies and proposes the concept of performance predictor. Based on this, it introduces the latest research achievements of our group in the field of performance predictors. In addition, this speech also introduces the latest progress of our group using NAS technology in the fields of robust machine learning and continual learning, demonstrating the broad potential and future development direction of NAS technology.
Biography:
Yanan Sun is currently a Professor with the College of Computer Science, Sichuan University, a selected candidate for the national youth support plan, and a selected candidate for the Tianfu Emei support plan. His research interests include evolutionary computation, neural networks, and their applications on neural architecture search. Dr. Sun serves as an Associate Editor for IEEE TRANSACTIONS ON EVOLUTIONARY COMPUTATION and IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS. In the past five years, he has published an authored book, and 50 papers as the first/corresponding author in IEEE Transaction, CCF-A, etc. Eight papers were selected as ESI Highly Cited Papers and ESI Hot Papers, and three papers were awarded the Best Paper and Spotlight Paper of Artificial Intelligence International Conferences, with the highest number of citations of 706 times. He has obtained 25 Chinese and US invention patents and has undertaken 15 national and provincial research projects.
Address:Sichuan University, College of Computer Science and Engineering, Chengdu, China, 610054