Mark Heimann

95 posts

Mark Heimann

Mark Heimann

@lastgoodcaesar

Graph machine learning researcher @Livermore_Lab. @UMich ‘20 PhD @WUSTL ‘15 AB/MS. Chess master and powerlifter.

Katılım Ocak 2016
119 Takip Edilen185 Takipçiler
Mark Heimann
Mark Heimann@lastgoodcaesar·
A great combination of theoretical analysis, examination of intuition from other domains, empirical insights, and new benchmarks for graph contrastive learning—all in one paper. Outstanding work led by Puja!
Puja Trivedi@puja_computes

New Paper Alert #NeurIPS2022: Self-supervised learning with discrete, structured data comes with some twists! In our paper, we explore how data-centric properties known to be crucial to success of visual CL can fail to hold for graph CL. 🧵Abs:arxiv.org/abs/2208.02810

English
0
1
15
0
Mark Heimann
Mark Heimann@lastgoodcaesar·
@d_metaxa Hey, just wanted to say, as another early career CS researcher and powerlifter (also newbie climber!), you might enjoy Collective Strength. They have competition-grade PL equipment and an admirable ethos collectivestrengthphl.com
English
0
0
0
0
Mark Heimann
Mark Heimann@lastgoodcaesar·
@rjurney Closest thing I know of is a new conference at the end of this year that's devoted to machine learning on graphs: #hero" target="_blank" rel="nofollow noopener">logconference.org/#hero
English
0
0
2
0
Russell Jurney 🇺🇦
Russell Jurney 🇺🇦@rjurney·
I want to attend a motif and graphlet machine learning conference - has there ever been such a thing?
English
1
0
0
0
Mark Heimann
Mark Heimann@lastgoodcaesar·
@rtcraig3 To be precise, this is the highest form of grad school
English
0
0
1
0
Taylor Craig
Taylor Craig@rtcraig3·
Just submitted a paper between squat sets, is this grad school yet?
English
1
0
3
0
Mark Heimann retweetledi
Jiong Zhu (@jiongzhu@mastodon.social)
Thanks @tangjiliang for sharing this interesting and thought-provoking work. In light of these new findings, we revisit the problem of heterophily for GNNs and discuss the reasons behind the seemingly different takeaways from different works. Check it out: jiongzhu.net/revisiting-het…
Jiliang Tang@tangjiliang

GNNs are widely believed to work well due to the homophily assumption and recent works design new architectures to overcome such heterophily-related limitations. Is Homophily a Necessity for Graph Neural Networks? the answer is in our new preprint: arxiv.org/abs/2106.06134

English
3
2
13
0
Mark Heimann
Mark Heimann@lastgoodcaesar·
@QuestionofWhy I might also need to go to an actual barber for the first time in like 2 years lol
English
0
0
0
0
Mark Heimann
Mark Heimann@lastgoodcaesar·
@SarahJabbour_ Congrats, and glad to hear the weights are such a great outlet during your PhD :) I don't know how to do cool barbell movements but otherwise I relate
English
1
0
1
0
Mark Heimann
Mark Heimann@lastgoodcaesar·
@ElissaWelle I went through this too. So glad you're a) talking about it, and b) supporting others in similar places. And also glad you did stick around and produce some amazing work :)
English
0
0
3
0
Mark Heimann
Mark Heimann@lastgoodcaesar·
@rtcraig3 what a legend to have one's name be sandwiched between Aquinas and Kierkegaard
English
0
0
1
0
Taylor Craig
Taylor Craig@rtcraig3·
citations: Nathan Cartagena, Thomas Aquinas, Andrew Maxwell, and Kierkegaard
English
1
0
2
0
Taylor Craig
Taylor Craig@rtcraig3·
The goal of cultural engagement should be to turn philosophical water into theological wine, but instead we turn theological wine into culture-wars water; and often poisonous water at that.
English
1
2
8
0
Vagelis Papalexakis
Vagelis Papalexakis@vagelispapalex·
Super excited to let the world know that, as of July 1, I will be an Associate Professor with tenure at @UCR_CSE @UCRBCOE! Extremely grateful to everyone who has contributed to this, first and foremost my students, my collaborators, and mentors inside and outside of @UCRiverside
English
28
2
170
0
Mark Heimann
Mark Heimann@lastgoodcaesar·
@abuyukcakir There are so many worthwhile paths to take in life with or without a PhD. I look forward to seeing which of them you'll take next! The potential is truly unbounded
English
0
0
2
0
Alican Büyükçakır
Alican Büyükçakır@abuyukcakir·
Dear fellas, it is official now. I will be discontinued from my PhD program at the U of Michigan Ann Arbor as of August 29. It was a pleasure getting to know all of you. It was definitely weird times during my stay in the US with elections, BLM, Covid and GME pump-and-dump lol.
English
8
0
19
0
Mark Heimann
Mark Heimann@lastgoodcaesar·
@abuyukcakir Lol that guitar is yours to keep if you like! If not, of course it's a great idea to have it circle around the GEMS Lab a while longer (I hadn't even considered this but I love this idea too)
English
0
0
1
0
Alican Büyükçakır
Alican Büyükçakır@abuyukcakir·
Ummm closing comments... I will probably give @lastgoodcaesar's acoustic guitar to another member of GEMS Lab. Mark you are such a complex character man, a chess grandmaster and a deadlifter? Come on. Be mediocre at something. I will miss you too.
English
1
0
2
0
Mark Heimann
Mark Heimann@lastgoodcaesar·
arxiv.org/abs/2102.13582 Use node proximities to learn embeddings modeling node proximity OR structural roles. Kind of like using 🍕dough to make 🍕OR cinnamon rolls. @JingZhu85095487 did an outstanding job on this project and will present it today at #SDM2021
English
1
1
9
0
Mark Heimann
Mark Heimann@lastgoodcaesar·
@graph_ Nevertheless, congratulations on the defense of an impressive body of work for your PhD, Dr. Anton (oops, look what I did there, silly American that's me)
English
0
0
2
0
Anton Tsitsulin
Anton Tsitsulin@graph_·
The most annoying thing about German doctoral system is that I can not call myself Dr. Anton after the defence, it's only officially allowed after the library submission. Who needed to ruin the fun!?
English
5
0
13
0