2021年Blum教授主讲“有意识的图灵机”课程小结
新冠疫情期间,北京大学通过云端增进与各国大学、研究机构的合作,充分利用网络实现线上线下教育的深度融合,在“云端”重塑教育形态。
2020年春季,在北京大学“外籍访问学者前沿性本科英文课程”项目的支持下,图灵奖获得者、北京大学访问讲席教授、CMU 讲席教授 Manuel Blum 和夫人 CMU 杰出教授 Lenore Blum 通过网络平台首次开设“有意识的图灵机:认知与计算科学”短期课程,获30余名选修及旁听图灵班同学的一致好评。
2021年春季,该课程继续开课,Blum 教授夫妇还特别邀请了“家中机器学习方面的权威”、TTIC 教授 Avrim Blum 做客座报告。课程首次面向全校本科生开放部分选课名额。最终,近40位来自信息学院、元培学院、物理学院、生命科学学院的学生选修或旁听了课程并顺利结课。值得一提的是,有6位2020级学生在大一期间选修了该课程并以出色的成绩完成了各项作业。
新闻传送门:
图灵奖得主曼纽尔·布卢姆教授等两位学者北京大学聘任仪式暨学术讲座举行
2020年Blum教授主讲“有意识的图灵机”课程小结
关于课程
由于认知神经科学的重大进展,科学开始朝向了解大脑是如何实现意识这一方向发展。1988年,认知神经学家 Bernard Baars 教授提出了 Global workspace theory 和图灵的图灵机计算模型,旨在为人类大脑意识的产生提供一个简单而强大的数学模型。1990年,贝尔实验室的 Seiji Ogawa 等人发明了 fMRI,使神经科学家能够实时见证大脑实时工作。因此,曾经只有哲学家和神学家感兴趣的意识追求,现在开始被各学科的科学家所追求和探索。
这门短期课程主要介绍了三位 Blum 教授、也是该课程主讲人设计提出的有意识图灵机(Conscious Turing Machine,CTM)的理论模型,并探讨该形式的属性。
该课程属于图灵班前沿选修课,为期4周,每周2次、共8课时讲授,内容涵盖:
- The Global Workspace Model (GWM)
- The Turing Machine (TM)
- The Conscious Turing Machine (CTM)
- Special LTM Processors for Consciousness
- Pain and Pleasure: The Feelings of Consciousness
- What does consciousness do for computation?
从第二周开始,课程以2-3名同学自由组合完成小组作业,平均每次课一份作业,从最初的作业构想、计划,到中期的研究方法,再到最后形成正式的课程报告和小组汇报,分步骤进行。
考虑到时差原因,在上课之余,课程使用 Piazza 平台辅助教学,实现老师与同学随时的问答、沟通与互动。
注:Conscious Turing Machine 以下均简称“CTM”
Blum教授评语
Your students are stunning. We do not say this lightly. They managed to produce wonderful research papers - in an incredibly short three weeks! These papers have excellent ideas that we had not considered. It seems that Peking University undergraduates perform at the level of top PhD students at the best universities. We are truly astonished. Amazed !!!
——Manuel & Lenore Blum
助教有话说
非常荣幸能够和 cyr 学姐共同担任本次 Conscious Turing Machine 课程的助教,这是一次难得的机会,也是一项巨大的挑战。
首先,课程内容是一个全新的研究方向,使用计算机科学的方法对人的意识进行建模,这需要一些计算理论、机器学习等方向的基础,同时部分内容在去年的基础上又做了改动和优化,对同学们和我来说都是非常新颖的;在 Blum 教授深入浅出的讲授下,我们都收获颇多,同学们更是在短短一个月内做出了非常优秀的 project。
另外,今年的课程形式不同以往,大多数同学们在教室听,少数同学远程参与,Blum 教授在远程授课,对设备和组织都提出了较高的要求;在中心陈老师、李老师、杨老师提供的巨大帮助下,我们利用中心的网络会议系统顺利完成了一个月的课程,让 Blum 教授看到每一位同学,让每一位同学都能顺利参与到课堂讨论中。
最后,感谢 Blum 教授的精彩课程,感谢 CFCS 提供的宝贵机会。去年我是课程的受益者,今年有更多地同学从课程中收益,也希望未来能够有更多图灵奖得主、优秀计算机科学家和图灵班同学之间交流的机会。
——助教X同学
结业合照,Blum教授夫妇专门打印并标注,确保对应上每一位同学
再次担任 CTM 这门课的助教,在听课的过程中,发现 Conscious Turing Machine 的理论和去年相比,又有了新的内容。虽然线上与线下结合的上课模式给师生互动与教学都造成了不小的挑战,两位教授的教学水平还是不受影响,同学们也非常积极、热情地学习和提问,还有来自瑞典的小伙伴提供了新的角度和思考。在协助课堂教学的同时和大家一起学习,发现自己的成长,非常感谢这次的助教机会!
——助教C同学
课程感受
这门课的内容很有趣,通过这次课程我学到了不少关于 CTM 的知识,也通过课程实践对 CTM 的理论基础和数学模型有了更为深入的知识,感谢两位教授和 TA 的付出!
——Z同学
三位 Blum 教授和同学们都很有创造力,让我在这一个月接触、学习到了很多有趣的东西。自己参与工作的 project 也得到了大家的认可与鼓励,感觉特别棒。
——H同学
It's been a wonderful journey for me! I really like how the course is organized.
——Y同学
This course stimulated my interest in TCS, neuroscience, and even philosophy.
——S同学
The model professors introduced in this course is surely an exemplar of interdisciplinary research which touches upon computer science, maths, neuroscience, behavioral psychology and social science. Still there are many intriguing topics to explore of this brand new model, and some concepts need further development. I am eager to follow up this ambitious work and make possible contributions in the future.
——D同学
This course is really meaningful to me because doing interdisciplinary research in neurosience and computer science is always my dream job, and CTM is definitely a good start point.
——H同学
The course not only gives insightful perspectives on consciousness, but also vividly shows how senior researchers gather their raw ideas, ask questions and refine the ideas, which is truly inspiring. Unforgettable experience indeed!
——W同学
Thanks to Blum family for bringing the creative and imaginative CTM course! This course not only deepened my understanding of human brain and consciousness, but also inspired me to continue to explore the field of Artificial Intelligence. We are looking forward to continuous improvement and more innovative achievements in CTM field.
——Y同学
The concept of CTM provides a novel interdisciplinary view to addressing the problem of consciousness, which is fundamental in understanding both the machine learning and human brain mechanisms. As deep learning has now come into the bottleneck, the integration of consciousness may endow artificial neural networks with a greater ability to generalize on unseen tasks. I am greatly inspired by this "consciousness architecture".
——C同学
This course is really fantastic! The model Conscious Turing Machine brought me not only a mathematical model of consciousness, but also a new way of thinking and questioning in my life.
——A同学
被Blum教授圈粉
Blum教授夫妇远程授课
Manuel Blum 和 Lenore Blum 教授都非常亲切和蔼,深入浅出的讲解方式给我留下了很深的印象。我原本以为意识是复杂而神秘的,没想到意识竟然能被这么简洁优美地建模。课程 project 也是一次有趣的科研训练;在我们对 project 没有信心时,两位教授总能从我们的 project 中发现亮点,给我们鼓励。感谢两位教授、TA 和同学们的全程陪伴!
——C同学
As a CS student, it's such a great and rare opportunity to be able to be instructed by Turing award recipients, especially with such deep and interesting insights into consciousness.
——W同学
This course is not merely a course on a new theoretical computer science model. Rather, I can feel from the lectures the Blums' passion for this new model.
——W同学
Thank you Professor Lenore and Manuel Blum for your wonderful guidance. And it's very nice of you to even stay up late to give detailed comments on our project.
——L同学
I've also learned a lot from our dear professors, in how we can define and analyze a problem (and winning a Turing Award, I wish).
——A同学
What a fantastic journey into the world of CTM! Thank you, all three Blums, for your wonderful talks, your valuable feedback that guided us all the way through the project and your effort to overcome such inconvenience caused by jet lag and unstable reception!
——D同学
Thanks so much for Blums' wonderful course. I have learnt a lot from their spirit of keep exploring new and unknown research topics.
——W同学
Great thanks to Blum Family for such an inspiring and outstanding course. The insightful CTM model indicated a TCS way towards 'consciousness'.
——S同学
I have learned a lot from this course, not only because Blum's insightful model of CTM inspired us a lot, but also for their kindness, patience, and their passion towards research.
——L同学
Thanks professors for bringing to us such a wonderful course! We talked about many ideas and got many inspirations from professors and students.
——H同学
Mr. and Mrs. Blum are so nice and patient that they carefully looked through every sentence I hand in, pointing out mistakes and giving advice. Thanks to the professors and TAs for giving us such a great course!
——L同学
课程意见
- 课程时间还是太短呀
- 课程时间与部分 ddl 高度重叠
- 设施(如麦克风、音响等)还有提高的空间,以保证更流畅的课堂互动
- 希望疫情结束能邀请 Blum 教授一家来北大上课
~以上意见都已收到~
部分课程作业题目
- Classical Cognitive Phenomenon With CTM Explanations Ye Yuan, Yunzhe Liu, Pu Yi
- CTM: A Game Theory Perspective Hanyue Lou, Haojun Yu, Zhuming Shi
- CTM Application in Classic Game Theories Hao Wang, Lucas Wang, Yuechen Zhu
- DoorWay CTM: Reproduction and Explanation Lan Lyu, Bowen Su, Chang Ma
- How CTM Designs and Writes Programs Wenyang He, Yu Chen, and Boshi An
- How CTM Explains Dreams Chuwen Liu, Hanyu Liu
- Mathematical Foundations of CTM Lingxiao Zheng, Rui Ding
- Model-of-the-World Processorsin a Compositional and Adversarial Manner Guanqi Zhan, Bi'an Du, Daichao Chen
- MusiCTM: On Music Composing by CTM Zongyuan He, Shuchen Li, Chang Wang