時間：114年05月26日(星期一) 19:00-21:00

地點： 大仁樓301

主持人：紀明德老師

演講者: 臺灣師範大學 王科植副教授

演講題目： Machine Learning for Ensemble Scientific Data Processing and Visualization

演講摘要： With the exponential growth in scientific simulations, the volume and

complexity of ensemble datasets have introduced major challenges in data storage,

I/O latency, and real-time analysis. This talk presents an AI-augmented pipeline for

large-scale scientific data processing and visualization, emphasizing the synergy

between in situ computation, data proxies, and deep learning techniques. By

integrating machine learning models into the simulation-visualization workflow, we

demonstrate how to alleviate storage bottlenecks and reduce computation overhead

while preserving essential data characteristics. We present a set of representative

systems that utilize deep learning techniques to facilitate visualization during

simulation, interactive parameter exploration, and uncertainty-aware data

reconstruction. This talk concludes with a discussion of the implications of AI in

scientific workflows. These advancements enable scalable, interactive, and insightful

exploration of complex ensemble simulations in diverse scientific domains.

演講題目: Recent Topics on Cryptography

演講綱要:

1) CIA and Application Scenarios

2) User Authentication vs Device Authentication

3) Cyberattacks vs Physical attacks

4) Classical Attacks vs Quantum Attacks

5) Classical cipher vs Quantum Cipher vs Post-Quantum Cipher

6) Discussion on an Anti-Quantum Key Exchange Protocol