ECE 209AS: AI/ML Foundations for IoT and Human-Cyber-Physical Systems

Hi All:

I've put 30-min project presentation slots per team for Wed March 17  at

https://doodle.com/poll/w25frxp6gb7shr6r?utm_source=poll&utm_medium=link

Please sign up - only one person per team should sign up. If perchance none of these slows can work for you, then please discuss: I have availability on next week's Monday and Tuesday as well.

Also, instructions for slide preparation at 

https://docs.google.com/document/d/1ncv4AeZf5yDXf9EnQotevXIPRD0f51XNacSTaxnHUEs/edit?usp=sharing

Lastly, if a demo makes sense, you should do as well (or can have a recording of that which you can show as part of your presentation).

Mani

Hi Everyone:

Yesterday I added to the schedule the final two papers for the course.

• Paper #9
  Gan, Chuang, Hang Zhao, Peihao Chen, David Cox, and Antonio Torralba. "Self-supervised moving vehicle tracking with stereo sound." In Proceedings of the IEEE International Conference on Computer Vision, pp. 7053-7062. 2019.
  https://openaccess.thecvf.com/content_ICCV_2019/papers/Gan_Self-Supervised_Moving_Vehicle_Tracking_With_Stereo_Sound_ICCV_2019_paper.pdf
  Presenter: WANG, MINGYU

• Paper #10
  Yao, Shuochao, Ailing Piao, Wenjun Jiang, Yiran Zhao, Huajie Shao, Shengzhong Liu, Dongxin Liu et al. "Stfnets: Learning sensing signals from the time-frequency perspective with short-time fourier neural networks." In The World Wide Web Conference, pp. 2192-2202. 2019.
  https://dl.acm.org/doi/pdf/10.1145/3308558.3313426
  Presenter: YOUNG, GRANT CHARLES

Hi everyone:

Since the changes in schedule caused some confusion as to when reviews of Paper #8 (which Swapnil is presenting today) would be due, I have extended the submission deadline for it to 10AM tomorrow. The paper would still be presented today, and please do try to do at least a rapid read of it prior to the class.

Mani

Hi Everyone:

In light of my unforeseen cancellation of yesterday's lecture, I'd like to do the following change in plan for the remaining classes that I feel would be the easiest way to make things up.

1. Thu March 4: wrap up current slide set (+ 2 papers from Tue 3/2 + 1 paper scheduled for Thu 3/4

2. Tue March 9: full lecture

3. Thu March 11: short lecture + 2 papers originally scheduled for Tue 3/9

4. Instead of in-class final project presentations that were scheduled for March 11, I'll provide a bunch of slots during the finals week for each team to present to me - each slot will be mandatory for the presenting team but open for anyone in the class to attend depending on availability and interest. This will also allow for a more relaxed presentation as I feel the mid-term presentations came under time pressure, and also give you more time to prepare for the final presentation.

Mani

Hi Everyone:

As it turns out, I need to go in tomorrow to doctor for a medical issues that has cropped up and needs urgent attention. There is a slight chance that I may not be able to be back in time for the lecture, and also not be in a state to communicate if undergoing any test/procedure. So please come to the lecture but if you don't see there by a few min after 2pm then assume that I am stuck. If that happens, we will shift the paper presentation to Thu, and I will do a lecture recording and distribute. 

Sorry about this and thanks for understanding!

Mani

Hi Everyone:

Below are the paper assignment for the next week.

Mani

Tuesday  March 2:

• Paper #6
  Ma, Meiyi, Ji Gao, Lu Feng, and John Stankovic. "STLnet: Signal Temporal Logic Enforced Multivariate Recurrent Neural Networks." Advances in Neural Information Processing Systems 33 (2020).
  http://www.cs.virginia.edu/~mm5tk/papers/NeurIPS20_STLnet.pdf
  Presenter: MOURE, PEHUEN PATRICIO

• Paper #7
  Yao, Shuochao, Jinyang Li, Dongxin Liu, Tianshi Wang, Shengzhong Liu, Huajie Shao, and Tarek Abdelzaher. "Deep compressive offloading: speeding up neural network inference by trading edge computation for network latency." In Proceedings of the 18th Conference on Embedded Networked Sensor Systems, pp. 476-488. 2020.
  https://dl.acm.org/doi/pdf/10.1145/3384419.3430898
  Presenter: NICHOLAS, MATTHEW CARTER

Thursday March 4:

• Paper #8
  Cai, Han, Chuang Gan, Ligeng Zhu, and Song Han. "TinyTL: Reduce Memory, Not Parameters for Efficient On-Device Learning." Advances in Neural Information Processing Systems 33 (2020).
  https://hanlab.mit.edu/projects/tinyml/tinyTL/assets/tinyTL_paper.pdf
  Presenter: SAHA, SWAPNIL SAYAN

Dear Students:

Please note that next Tuesday, Feb 23 we have the project midterm presentations. The instructions are at https://docs.google.com/document/d/15DtIOJ9RTrpBl4dfkGVC68BWqVGW7zW0zouziQtNJ6o.  Also, I have put the list of projects at https://piazza.com/class/kjhpcsjab6d72j?cid=38. We will do the presentations in the order there.

Lastly, I see that several of the project websites are woefully incomplete relative to the instructions in my prior posting a couple of weeks ago (https://piazza.com/class/kjhpcsjab6d72j?cid=22). Please note that your project proposal score depended on the content of the website. In some cases, all I see is one sentence and in some others, just a tiny bit with no evidence whatsoever that you planned any specific aims, did any study of related work, etc. While the proposal deadline is long past, it is still important that you update it by the midterm presentation date. Please remember, the website *is* your report and it needs to be continually updated.

Mani

Below are the papers for Tuesday March 2

• Paper #6
  Ma, Meiyi, Ji Gao, Lu Feng, and John Stankovic. "STLnet: Signal Temporal Logic Enforced Multivariate Recurrent Neural Networks." Advances in Neural Information Processing Systems 33 (2020).
  http://www.cs.virginia.edu/~mm5tk/papers/NeurIPS20_STLnet.pdf
  Presenter: MOURE, PEHUEN PATRICIO

• Paper #7
  Yao, Shuochao, Jinyang Li, Dongxin Liu, Tianshi Wang, Shengzhong Liu, Huajie Shao, and Tarek Abdelzaher. "Deep compressive offloading: speeding up neural network inference by trading edge computation for network latency." In Proceedings of the 18th Conference on Embedded Networked Sensor Systems, pp. 476-488. 2020.
  https://dl.acm.org/doi/pdf/10.1145/3384419.3430898
  Presenter: NICHOLAS, MATTHEW CARTER

Mani