@MoeinShariatnia Profile picture

Moein Shariatnia

@MoeinShariatnia

ML Dev & Researcher | 🩺 Med Student at TUMS | CV and NLP | Writer @YuanCommunity and @Medium | Freelancer AI dev

Similar User
MayoAILab photo

@MayoAILab

Elham Mahmoudi, MD, MPH photo

@EliMahMD

Niloofar Motamedi, MD photo

@NiloofarMotame2

Kasper Green Larsen photo

@kasperglarsen

Homa Dadras, MD photo

@Homa_Dadras

Shikhar photo

@ShikharMurty

Mahdie Hosseini, MD photo

@MahdieHosseini6

Ben Eysenbach photo

@ben_eysenbach

Pouria Rouzrokh, MD, MPH, MHPE photo

@PRouzrokh

ManaMoassefi photo

@ManaMSF94

Minji Yoon photo

@MinjiYoon90

Negin Badihian, MD photo

@NBadihian

Fereshteh Yazdanpanah, MD, HMBA photo

@FYazdanpanah90

Mahsa Dolatshahi photo

@MahsaDolatshah1

Bardia Khosravi, MD photo

@BrdKhsrv

Pinned

📄🔥 Check out our latest work on "The Role of Pre-training Data in Transfer Learning". We investigate a few interesting questions regarding the dataset #distribution, #noise, #size, and various training #methods; and how they impact #transfer learning. arxiv.org/abs/2302.13602

Tweet Image 1

🆕 📰 The Role of Pre-training Data in Transfer Learning You have a target task and wonder which pre-training model from @huggingface to use for transfer learning? arxiv.org/abs/2302.13602 w @Mitchnw @osaukh @MoeinShariatnia @HanieSedghi @lschmidt3 1/8



Moein Shariatnia Reposted

🎉 For the past couple months, I've been keeping myself busy with something exciting — meet Najva! A free native macOS app that brings together on-device transcription and your favorite AI models, right from your menubar. 🧵👇


Moein Shariatnia Reposted

How to achieve success without trying to shoot for it! My notes from the #book "Why Greatness Cannot be Planned" by @kenneth0stanley and @joelbot3000 🧵

Tweet Image 1

Moein Shariatnia Reposted

Want a really simple (and cheap!) way to improve neural net calibration and get practical epistemic uncertainty estimates? At ICLR this year: Variational Bayesian Last Layers Try it out: 1. pip install vbll 2. a couple one line changes to your current training pipeline 🧵

Tweet Image 1

Moein Shariatnia Reposted

“What did you get done this week?”


Moein Shariatnia Reposted

Just came across a super-innovative paper by @MassGenBrigham where the authors tokenized all events happening to a cohort of patients and trained a GPT model to predict the next token given the prior ones. It’s like turning a patient’s clinical course into a language that we only…

Tweet Image 1

Moein Shariatnia Reposted

We put OpenAI o1 to the test against ARC Prize. Results: both o1 models beat GPT-4o. And o1-preview is on par with Claude 3.5 Sonnet. Can chain-of-thought scale to AGI? What explains o1's modest scores on ARC-AGI? Our notes: arcprize.org/blog/openai-o1…

Tweet Image 1

How it started vs How it's going :) #o1 #openai

Tweet Image 1
Tweet Image 2

Moein Shariatnia Reposted

AI is incredible at writing code. But that's not enough to create software. You need to set up a dev environment, install packages, configure DB, and, if lucky, deploy. It's time to automate all this. Announcing Replit Agent in early access—available today for subscribers:


Moein Shariatnia Reposted

Last Week in Medical AI: Top Research Papers/Models 🏅(August 17 - August 24, 2024) - Jailbreak on Medical Multimodal LLMs - LLMs are not Zero-Shot Biomedical Reasoners - RuleAlign framework: Aligning LLM for Physician Rules - CTP-LLM: LLMs for Clinical Trial Transition…

Tweet Image 1

Can't agree more!

Biologists taking machine learning techniques from computer vision and nlp

Tweet Image 1


Moein Shariatnia Reposted

We have pushed a new update to the GDL book chapters 📐🌐🤖 Join Alice on her journey to the Geometric wonderland! * Chapters 2&3 (first 100 pages!) ✍️ * Companion slides! 💬 * [By popular demand] _Exercises_! 💪 geometricdeeplearning.com/book 📖 @mmbronstein @joanbruna @TacoCohen

Tweet Image 1
Tweet Image 2

Moein Shariatnia Reposted

If you remember our Applied LLMs course, you'll love this. Today, we are making all these resources available for free to everyone! 📚 We did extra work to add learning tracks, resources, and notes to each lesson to maximize your learning. Link in next tweet

Tweet Image 1

Just passed 100 citations on my publications! scholar.google.com/citations?user…

Tweet Image 1

Moein Shariatnia Reposted

Our EEML Startup Panel is well underway! 🚀Featuring insightful advice from our stellar panelists -- all with ties to the EE region: @mmbronstein @GeorgeRedTech @talexgeorge @nikola_mrksic Ana-Maria Constantin and Jelena Zorić. Kindly moderated by @gnutekm 🥳

Tweet Image 1

Moein Shariatnia Reposted

Impressive talk by @redpony! 🥳🥳🥳

Next up at EEML'24: @redpony giving an iconic talk on fine-tuning LLMs 🔧 through SFT and RLHF 💰

Tweet Image 1


It feels great to finally meet your online #DL / @PyTorch tutor in-person! Thanks @alfcnz for all the effort you put in making amazing videos and presentations for the community. @EEMLcommunity #EEML

Tweet Image 1

Moein Shariatnia Reposted

Starting @EEMLcommunity 2024 with an outstanding @PetarV_93 Of course, I’m around as well, representing @NYUDataScience and @NYU_Courant 😇😇😇

Tweet Image 1
Tweet Image 2

The official start of Eastern Europe Machine Learning summer school #2024 With the legendary @PetarV_93

Tweet Image 1

Loading...

Something went wrong.


Something went wrong.