Baharan Mirzasoleiman (@baharanm) / X

Baharan Mirzasoleiman

78 posts

Baharan Mirzasoleiman

@baharanm

Assistant professor @UCLAComSci. Better ML via better data, Machine learning, Optimization

Los Angeles, CA

web.cs.ucla.edu/~baharan/

Joined July 2018

Pinned
Baharan Mirzasoleiman
@baharanm
Aug 3, 2024
We’re thrilled by the amazing response to our #ICML2024 tutorial on “Foundations of data-efficient learning”! Over 1000 attendees joined us. Thank you all! 🙌🌱🌱🌱 ➡️ Slides: baharanm.github.io/ICML24_tutoria… ➡️ Recording: will be available on Aug 22 🎊🎊
Baharan Mirzasoleiman
@baharanm
Jul 19, 2024
I'll be giving a 2-hour tutorial on data-efficient learning with my PhD student @sjoshi804 on Monday July 22 at #ICML2024. Join us to learn more about this cool topic! ➡️ We can learn better from better data! ⬅️🙌🌱
Baharan Mirzasoleiman
@baharanm
Jul 19, 2024
I'll be giving a 2-hour tutorial on data-efficient learning with my PhD student @sjoshi804 on Monday July 22 at #ICML2024. Join us to learn more about this cool topic! ➡️ We can learn better from better data! ⬅️🙌🌱
Baharan Mirzasoleiman
@baharanm
Jul 19, 2024
ML models are sensitive to distribution shift. Can we adapt a model with only a few examples from the target domain? In this #ICML2024 paper, @xue_yihao65785 proposes an effective way, with nice theoretical analysis🌱 🔗arxiv.org/pdf/2305.14521 Thu, July 25, Poster session 5, #800
Baharan Mirzasoleiman
@baharanm
May 5, 2023
One of our recent work that I’m most excited about: speed up your large (contrastive) pretraining by 2.5x without compromising accuracy! This can be done (rigorously) by pretraining on the “right” data points! Shout out to @sjoshi804 for his amazing work accepted to #ICML2023!
Siddharth Joshi
@sjoshi804
May 4, 2023
Our paper (@baharanm) Data-Efficient Contrastive Learning sjoshi804.github.io/data-efficient… has been accepted to #ICML2023! We propose the first method to select the most useful data subset to train on and provide rigorous theoretical generalization guarantees. 🧵 (1/5)
Baharan Mirzasoleiman
@baharanm
Apr 16, 2024
My PhD student @xue_yihao65785 has received one of the 50 @OpenAI Superalignment Fast Grants (out of 2700 applications)! Big congrats Yihao and looking forward to seeing more amazing work from you! 🎉🎉🌱🌱
Yihao Xue
@xue_yihao65785
Apr 10, 2024
Huge thanks to @OpenAI for the superalignment fast grants! Can't put into words how happy I am. Looking forward to diving into a seriously exciting project!
Baharan Mirzasoleiman
@baharanm
Dec 13, 2024
I’ll help presenting our #NeurIPS2024 posters tomorrow (Friday):🌱 1- Changing the training data distribution to improve in-distribution performance (11@west #7106) w. @dangnth97 2- Data selection for fine-tuning LLMs with superior performance (16:30@west #5401) w. @YUYANG_UCLA
Baharan Mirzasoleiman
@baharanm
Mar 4, 2025
Big congrats @YuYang_i on your graduation!! 🎉🎉 🎉 very nice PhD thesis with great contributions 🌱 I’m proud of all you’ve done, and I wish you the best! 💝
Yu Yang
@YuYang_i
Mar 1, 2025
Sharing a little late update (before it’s no longer news): I wrapped up my PhD at the end of last year and recently joined @OpenAI’s reasoning team 🍓✨!
Baharan Mirzasoleiman
@baharanm
Dec 11, 2024
Smaller high-quality subsets of language data not only improve LLMs’ training efficiency, but also yield considerably better performance! 🙌🎉🌱 @YUYANG_UCLA has a theoretically-rigorous method for this in her #NeurIPS2024 paper! Check it out on Fri, Dec 13, 16:30, #PS 6
Yu Yang
@YuYang_i
Dec 9, 2024
1/ I'll be at #NeurIPS2024 presenting our work SmallToLarge (S2L): Data-efficient Fine-tuning of LLMs! 🚀 What’s S2L? It’s a scalable data selection method that trains a small proxy model to guide fine-tuning for larger models, reducing costs while preserving performance. 👇
Baharan Mirzasoleiman
@baharanm
Jul 19, 2024
CLIP is highly sensitive to data poisoning and backdoor attacks. In this #ICML2024 paper, @WenhanYang0315 proposed an interesting way to pretrain CLIP robust to such attacks without compromising the performance! 🌱🌱 🔗arxiv.org/pdf/2310.05862 Thu, July 25, Poster session 6, #814
Baharan Mirzasoleiman
@baharanm
May 6, 2024
Why CLIP is more robust to distribution shift than supervised learning? This #ICLR2024 paper provides the first rigorous proof! TL;DR details specified in the captions allow learning more generalizable features from images. Check it out: Tue, PS#1, P#113 arxiv.org/pdf/2319.04971
Yihao Xue
@xue_yihao65785
Mar 20, 2024
Happy to share that I have two papers (arxiv.org/pdf/2310.04971… , arxiv.org/pdf/2403.11391… ) accepted at #ICLR2024! ⬇️🧵 1/
Baharan Mirzasoleiman
@baharanm
Jul 17, 2024
Double descent confirms the benefit of larger models. But, when there is label noise in the data, larger model size can hurt the performance! We called this phenomenon "Final Ascent". Check out this interesting #UAI2024 spotlight by @xue_yihao65785: arxiv.org/pdf/2208.08003 🙌🌱
Baharan Mirzasoleiman
@baharanm
Jun 14, 2023
Can we speedup deep learning 2.5x via data-selection with theoretical guarantees? Our #ICM2023 paper shows for the first time that this is indeed possible! Data points of increasing learning difficulty is what you need! An excellent work well worth reading (@YUYANG_UCLA,Hao)🎉🌱
Yu Yang
@YuYang_i
Jun 14, 2023
📢Our (Hao Kang, @baharanm) paper (arxiv.org/abs/2306.01244) will appear at #ICML2023! Introducing CREST: the first coreset selection algorithm theoretically guaranteed to speed up training of deep neural networks!🚀(🧵1/7)
Baharan Mirzasoleiman
@baharanm
Jun 23, 2023
Neural networks can make right predictions for wrong reasons! Despite the body of efforts on mitigating spurious correlations,this problem is far from being solved on real-world datasets. We introduced a benchmark and two datasets to help you evaluate and develop better methods🎉
Siddharth Joshi
@sjoshi804
Jun 23, 2023
Introducing SpuCo: a Python package to standardize tackling spurious correlations in deep neural networks! 1️⃣ Modular implementation of current SOTA methods 2️⃣ Two new challenging and realistic datasets Paper: arxiv.org/abs/2306.11957 Github: github.com/BigML-CS-UCLA/… 🧵(1/n)
Baharan Mirzasoleiman
@baharanm
Apr 24, 2025
Can we pretrain deep models with small synthetic data? Dataset Distillation via Knowledge Distillation is the way to go! Check out @sjoshi804’s #ICLR2025 paper this Saturday April 26 at 9am, Poster #307 🎉🌱
Siddharth Joshi
@sjoshi804
Apr 23, 2025
#ICLR2025 Can you pre-train deep models with small, synthetic datasets? 🤯 We introduce the first effective dataset distillation method for self-supervised learning (SSL) — boosting downstream accuracy by up to 13% over baselines. 🧪 Poster #307, Sat Apr 26, 9am