@counterfac

Assistant Professor @PurdueECE | Postdoc @MIT_CSAIL | PhD @UChicagoCS

 West Lafayette, IN
y-ding.github.io/
Joined July 2013
  • Tweets110
  • Following580
  • Followers333
6Media
Save the Date for an #NSF DESC Workshop on Sustainable Computing on Aug 20-21, 2024. This workshop will focus on the dimensions of #AI, water, and biodiversity dimensions, extending beyond just carbon footprints. #SustainableComputing
1
2
529
I am recruiting PhD students at Purdue ECE (not CS) starting from Fall 2024! We have ongoing projects in: - Observational Studies and Causal Inference for Distributed Systems. - Sustainable Cloud Datacenters Students with background in sys, arch, ML are welcome to apply!
1
20
60
7,362
Application link: engineering.purdue.edu/ECE/A… Deadline: December 1, 2023 #phdopportunity #systemsresearch #comparchresearch #mlsys

Graduate Programs : Academics - Elmore Family School of Electrical and Computer Engineering -...

Purdue University's Elmore Family School of Electrical and Computer Engineering, founded in 1888, is one of the largest ECE departments in the nation and is consistently ranked among the best in the...

engineering.purdue.edu
1
476
Yi Ding retweeted
NITRD@NITRDgov25 May 2022
#NITRD30th has posters presented by @NSF & @compcomcon's Computing Innovation Fellows! Take a look @ CIFellows' cutting-edge research on emerging #NIT technology. What a great career-enhancing bridge experience NSF & Computing Community Consortium provide nitrd.gov/30th-anniversary-o…

NITRD 30th Anniversary Symposium: Poster Session

Backed by funding from the National Science Foundation, the Computing Research Association (CRA) and the Computing Community Consortium (CCC) announced a Computing Innovation Fellows (CIFellows)...

nitrd.gov
3
1
Attending @NITRDgov's #NITRD30th event organized by @compcomcon at the @IntlSpyMuseum #CIFellow
1
9
🚨New paper🚨 at Onward! “Programming with Neural Surrogates of Programs” with @counterfac and @mcarbin We show why and how you should replace all your programs with neural networks. alexrenda.com/onward-2021 youtube.com/watch?v=rM8SwKOB… 🧵1/13
3
20
5
98
The neuro-symbolic learning/programming space is quite varied in its ideas. But, this is neat work with the team cleaving out a survey and demonstration of a narrower set of work on replacing programs with neural networks, along with what you get (and don't!) when you so do.
Alex Renda@alex_renda_20 Oct 2021
🚨New paper🚨 at Onward! “Programming with Neural Surrogates of Programs” with @counterfac and @mcarbin We show why and how you should replace all your programs with neural networks. alexrenda.com/onward-2021 youtube.com/watch?v=rM8SwKOB… 🧵1/13
2
16
How to incorporate prior knowledge and understand how fast software configurations are arrived? Our paper: Generalizable and Interpretable Learning for Configuration Extrapolation (w/ A. Pervaiz, H. Hoffmann, @mcarbin) published in #esecfse21 @FSEconf, has a few ideas!
1
6
1
22
Paper: y-ding.github.io/papers/FSE2… We introduce a new configuration extrapolation tool, GIL+. GIL+ builds a hierarchical performance model by adding a low-level system metrics layer between SW configurations and final performance.
1
2
With the extra layer, GIL+ produces interpretable results to help users understand the underlying SW/HW factors that cause the low performance, and gain insights into how faster configurations are achieved.
2
Excited to see this work out. Great collaboration on applying causal DAG analysis as the quantitative tool to support findings on the relationships between park visits and crimes. 👇
Kate Schertz@kate_schertz27 Apr 2021
Excited to announce our new paper: Neighborhood street activity and greenspace usage uniquely contribute to predicting crime with @jamessaxon @BettencourtLuis @kharloews @counterfac Hank Hoffman @EnvNeuroLab Marc Berman. @Nature_NPJ Urban Sustainability: nature.com/articles/s42949-0…
3
I updated my list of causal system papers in #PL #SE column. Thank @jimmykoppel for suggesting these amazing papers! It seems PLSE is ahead of other CS fields in incorporating causality. Dependence is the key. github.com/y-ding/causal-sys…

GitHub - y-ding/causal-system-papers

Contribute to y-ding/causal-system-papers development by creating an account on GitHub.

github.com
Yi Ding@counterfac28 Jan 2021
Tired of ML for Systems? It's time to rethink and get a perspective on causal inference for systems. I assembled a paper list on Causal Systems, covering areas of #architecture, #database, #networking, #OS, #SE, and #PL: 👇 github.com/y-ding/causal-sys…

GitHub - y-ding/causal-system-papers

Contribute to y-ding/causal-system-papers development by creating an account on GitHub.

github.com
1
4
Thanks CRA @CRAtweets to feature my CIFellow project on Improving System Efficiency and Reliability with Causal Learning! Check it out👇
This Post is from an account that no longer exists. Learn more
1
10
Thank CCC @compcomcon for featuring my project on Improving System Efficiency and Reliability with Causal Learning! Check it out👇
This Post is from an account that no longer exists. Learn more
1
1
1
4
Link: cccblog.org/2021/03/18/cifel…
Interested in robust ML models other than robust neural networks? Check out robust ensemble tree 👇
Yizheng Chen@surrealyz28 Feb 2021
Using Twitter spam detection as an example, I wrote about how to train robust trees for security (USENIX Sec 21). Our most exciting result is, we can increase the feature manipulation cost for adaptive attackers to evade the robust tree ensemble by 10.6X. surrealyz.medium.com/robust-…

Robust Trees for Security

Tree models are widely used for security, such as detecting malicious autonomous system, social engineering, malware distribution…

surrealyz.medium.com
1
8
Tired of ML for Systems? It's time to rethink and get a perspective on causal inference for systems. I assembled a paper list on Causal Systems, covering areas of #architecture, #database, #networking, #OS, #SE, and #PL: 👇 github.com/y-ding/causal-sys…

GitHub - y-ding/causal-system-papers

Contribute to y-ding/causal-system-papers development by creating an account on GitHub.

github.com
5
8
3
26
It is inspired by System Configuration Management list from @tianyin_xu xitter.cf/tianyin_xu/status/1338…
Tianyin Xu
@tianyin_xu14 Dec 2020
Replying to @tianyin_xu
I assembled a list of papers and articles on configuration management for cloud/systems: github.com/tianyin/configura… Hope it helps.

GitHub - tianyin/configuration-management-papers: A Reading List of System Configuration Management

A Reading List of System Configuration Management. Contribute to tianyin/configuration-management-papers development by creating an account on GitHub.

github.com
1
1
In ML for systems problems #MLSys, is higher prediction accuracy equivalent to the better system outcome? The answer is NO! I wrote a blog post explaining why 👇: link.medium.com/7l1gRg5BTcb

Higher Prediction Accuracy != Better System Outcomes

This post shows that higher prediction accuracy is NOT equivalent to a better system outcome through a real example.

link.medium.com
1
1
7
I start an academic blog today and here is my first post: How to write good systems papers? Most of the tips come from my PhD advisor Hank Hoffmann. Retweets are appreciated. counterfac.medium.com/how-to…
1
18
2
63
@tianyin_xu @vj_chidambaram
1
1
Load more