IJTCS | 分论坛日程：机器学习与形式化方法
编者按
首届国际理论计算机联合大会（International Joint Conference on Theoretical Computer Science，IJTCS）将于2020年8月17日-21日在线上举行，由北京大学与中国工业与应用数学学会（CSIAM）、中国计算机学会（CCF）、国际计算机学会中国委员会（ACM China Council）联合主办，北京大学前沿计算研究中心承办。
本次大会的主题为“理论计算机科学领域的最新进展与焦点问题”。大会共设7个分论坛，分别对算法博弈论、区块链技术、多智能体强化学习、机器学习理论、量子计算、机器学习与形式化方法和算法与复杂性等领域进行深入探讨。同时，大会特别开设了青年博士论坛、女性学者论坛与本科生科研论坛，荟集海内外知名专家学者，聚焦理论计算机前沿问题。有关信息将持续更新，敬请关注！
本期带来“机器学习与形式化方法”分论坛精彩介绍。
“机器学习与形式化方法”介绍
近年来，深度学习技术越来越多地应用在无人驾驶、芯片等安全攸关领域。对于这类系统，传统测试方法已经无法满足人们提出的更高的系统可靠性要求。因此，机器学习和形式化方法领域交叉结合近年来成为研究热点之一，相关问题被越来越多的学者关注。本次会议讨论人工智能系统验证、测试技术，如何用形式化方法分析及提高人工智能系统可靠性等焦点问题，介绍国内外最新相关方法与技术。
“机器学习与形式化方法”分论坛主席
张立军
中国科学院
“机器学习与形式化方法”分论坛议程
时间：2020年8月19日
“机器学习与形式化方法”分论坛报告简介
孙 军
Neural Networks: Fairness and Interpretability
Abstract
In this work, I will present our recent work on analyzing neural networks in terms of fairness, i.e., solving the problem of checking whether a neural network inherits discrimination from real-world discriminative data. We start with a testing approach (i.e., falsificatoin) and subsequently work towards an approach for fairness verification. We show that discrimination often exists in neural network models and humbly seek your contribution to jointly address the problem.
薛 白
PAC Model Checking of Block-Box Continuous-Time Dynamical Systems
Abstract
In this talk I will present a model checking approach to finite-time safety verification of black-box continuous-time dynamical systems within the framework of probably approximately correct (PAC) learning. The black-box dynamical systems are the ones, for which no model is given but whose states changing continuously through time within a finite time interval can be observed at some discrete time instants for a given input. The new model checking approach is termed as PAC model checking due to incorporation of learned models with correctness guarantees expressed using the terms error probability and confidence. Based on the error probability and confidence level, our approach provides statistically formal guarantees that the time-evolving trajectories of the black-box dynamical system over finite time horizons fall within the range of the learned model plus a bounded interval, contributing to insights on the reachability of the black-box system and thus on the satisfiability of its safety requirements. The learned model together with the bounded interval is obtained by scenario optimization, which boils down to a linear programming problem.
黄小炜
Safety Certification of Deep Learning
Abstract
In the past few years, significant progress has been made in the development of deep neural networks (DNNs), which now outperform humans in several difficult tasks, such as image classification, natural language processing, and two-player games. Given the prospect of a broad deployment of DNNs in a wide range of applications, concerns regarding the safety and trustworthiness of this approach have been raised. In this talk, I will review two threads of our recent research towards safety certification of DNNs. The first thread is on the formal verification. We developed a few approaches that are able to provide provable guarantees to the verification of robustness and showed that they are able to work with real-world DNNs. The second thread of research is on the safety assurance of DNNs through testing-based method. This is an established approach that has been recommended in industrial standards such as ISO26262 for automotive software and DO178B/C for avionic software, and we adapted them for DNNs.
易新平
Spectral Analysis of Convolutional Neural Networks
Abstract
In this talk, I will present a spectral approach to analyze convolutional neural networks (CNNs). To understand the behaviors of convolutional layers in CNNs, we propose to use spectral density matrices to represent the linear transformation of convolutional layers. By doing so, spectral analysis of linear convolutional layers can be alternatively done on the corresponding spectral density matrices. Such a spectral representation will be demonstrated useful in obtaining e.g., singular value approximation and spectral norm bounding, which are employed as regularizers to enhance generalization performance in practical CNNs, e.g., ResNets.
关于IJTCS
简介 → 国际理论计算机联合大会重磅登场
推荐 → 大会特邀报告（一）
推荐 → 大会特邀报告（二）
日程 → 分论坛：算法博弈论
日程 → 分论坛：区块链技术
日程 → 分论坛：机器学习理论
日程 → 分论坛：量子计算
IJTCS注册信息
本次大会已经正式面向公众开放注册！每位参与者可以选择免费注册以观看线上报告，或是支付一定费用以进一步和讲者就报告内容进行交流，深度参与大会的更多环节。
观看线上报告：免费
完全注册：
（普通）$100 /￥700
（学生）$50 /￥350*
作为参会人参加全部会议，直接在线提问讨论并参与特设互动环节
注册截止：2020年8月15日23:59
点击 ↓↓↓二维码↓↓↓ 跳转注册页面：
*学生注册：网站上注册后需将学生证含有个人信息和学校信息的页拍照发送至IJTCS@pku.edu.cn，邮件主题格式为＂Student Registration + 姓名＂。
大会主席
John Hopcroft
中国科学院外籍院士、北京大学访问讲席教授
林惠民
中国科学院院士、中国科学院软件研究所专家
大会联合主席
邓小铁
北京大学教授
顾问委员会主席
高 文
中国工程院院士、北京大学教授
梅 宏
中国科学院院士、CCF理事长
张平文
中国科学院院士、CSIAM理事长、北京大学教授
组织单位
欢迎注册
大会网站：
https://econcs.pku.edu.cn/ijtcs2020/IJTCS2020.html
注册链接：
https://econcs.pku.edu.cn/ijtcs2020/Registration.htm
联系人
大会赞助、合作等信息，请联系：IJTCS@pku.edu.cn
— 版权声明 —
本微信公众号所有内容，由北京大学前沿计算研究中心微信自身创作、收集的文字、图片和音视频资料，版权属北京大学前沿计算研究中心微信所有；从公开渠道收集、整理及授权转载的文字、图片和音视频资料，版权属原作者。本公众号内容原作者如不愿意在本号刊登内容，请及时通知本号，予以删除。
点“阅读原文”转大会注册页面