Explore tweets tagged as #influencefunctions
@deep_thesis
DeepLearningThesis
7 years
Us debugging legacy influence function code. Hard at work or hardly working? #InfluenceFunctions #TensorFlow #DeepLearning
Tweet media one
0
0
0
@NicolaFitz
Nicola Fitz-Simon
4 years
Super talks this afternoon on recent research in #causalinference with #influencefunctions by @hines8 @podTockom @nshejazi @dscharf3 @TheEuroCIM hosted by @LSHTMstatmethod
Tweet media one
Tweet media two
Tweet media three
Tweet media four
2
3
8
Forgetting by Remembering: A Smarter Path to Machine Unlearning #machineunlearning #privacy #influencefunctions #incrementallearning #AIethics.
0
0
0
@LSHTM_datastats
Data & Statistical Science for Health, LSHTM
4 years
@NicolaFitz @hines8 @podTockom @nshejazi @dscharf3 @TheEuroCIM Nice choice of pics: pretty plots! We want to contribute this one to your collection #causalinference with #influencefunctions from our keynote speaker @edwardhkennedy
Tweet media one
1
0
6
@bryanklow
Bryan Kian Hsiang Low
5 months
🚀 Tired of slow, expensive data selection for training LLMs?. Still using #ShapleyValues or #InfluenceFunctions to find high-quality data, and burning through compute? 😩🔥💸. 🌟 The new @iclr_conf work of @xiaoqiang_98.@michael_xinyi on Efficient Top-m Data Values
1
2
13
@g_rpkg
gepuro task views
6 years
add {AnthonyChristidis/InfluenceFunctions} ( ) on
0
0
0
@deep_thesis
DeepLearningThesis
8 years
@botnet_hunter @cylanceinc Thanks for the suggestion! We read the paper and liked it so much that we posted a summary on our blog. Feel free to check it out, would appreciate any feedback! #InfluenceFunctions #DeepLearningThesis
0
0
1
@nanotrac
NanoTrac Technologies
2 years
Unlocking the Mysteries of Large Language Models: A Deep Dive into Influence Functions and Their Scalability | @Marktechpost #UnlockingMysteries #DeepDive #InfluenceFunctions #MachineLearning #AI #MarktechpostAI
0
5
0
@AZoAiNews
AZoAi
11 months
🔍🤖📊 Unlocking Transparency in Diffusion Models With Scalable Data Attribution Methods #AI #machinelearning #generativemodels #influencefunctions #diffusionmodels #datatransparency #scalability #research @arxiv.
0
0
0
@prernamishra92
Prerna Mishra
2 years
@toniwhited Pardon me if I sound idiotic. This is my first structural project and I am very nervous. I am following these noted: If I have 6 parameters to estimate and I simulated 10 panels, should I create influence functions for 6 parameters and *continued.
0
0
0
@jstatistic
Jay Kahn
6 years
@paulgp I actually have an old set of notes (badly in need of an update) on this too if it's of interest
0
0
5
@jstatistic
Jay Kahn
4 years
Went back to my influence function notes which I haven't looked at in six years, corrected some typos, and added some additional examples today. These are maybe the most visited thing on my website and I keep thinking I should do something with them.
0
0
5
@itsbradross
Brad Ross
4 years
@HazardYagan @toniwhited @instrumenthull You can just use calculus! A nice review of why is or for a more intuitive explanation that gives lots of examples, see
3
0
3
@jstatistic
Jay Kahn
4 years
@itsbradross @toniwhited @HazardYagan @instrumenthull This is the one? I've now noted that the stacking trick is from Erickson and Whited (2002). Sorry, I wrote these when I was a first-year to figure things out for myself, so there wasn't even a .bib file until today. But I should've done this earlier!.
0
0
0