Gökçen Büyükbaş

Explore how to fine-tune BERT using the Hugging Face Transformers library. Discover how you can leverage pre-trained models for your own projects from Shaw Talebi's article. #LLM #MachineLearning

13

Now available on edX, CS50's new Introduction to Programming with R, a popular language for statistical computing and graphics in data science and other domains, with CS50's own Carter Zenke. Overview at https://lnkd.in/eY-Rw4hk. Enroll for free at https://cs50.edx.org/r. #education #community #r #statistics 

1,584

34 Comments

Fine-tuning pre-trained models is a powerful paradigm for developing better models at a lower cost than training them from scratch. Check out Shaw Talebi's newest article now! #LLM #Python

35

Fine-tuning LLMs with Reinforcement Learning from Human Feedback This blog post explores how Reinforcement Learning from Human Feedback (RLHF) can be used to fine-tune Large Language Models (LLMs) to align with human preferences. It delves into the concept of Proximal Policy Optimization (PPO) and how it is used in this context. The key takeaway is that RLHF offers a powerful framework to train LLMs that can produce human-aligned text, while still maintaining creativity. This paves the way for more natural and nuanced language generation in the future of AI and natural language understanding. Read the article: https://lnkd.in/d9EGf579 #AI #machinelearning #goML

8

If you're not sure what the best way is to sample from multivariate distributions, don't miss Wencong Yang's helpful overview of a wide range of approaches, including kernel density estimation, GANs, and diffusion models.

5

Last week I was grateful to be invited to present our current working paper “Causal Inference with Cocycles” at the Munich Center for Machine Learning’s (MCML) 3rd annual workshop on causal machine learning. Causal machine learning goes beyond predicting outcomes based on patterns, and aims to identify actual causes behind those outcomes. 🧐In our work, we look at what most state of the art methods have in common, which is the need to (often) estimate entire distributions/densities from data. 🧠We show that in many cases, one can avoid this by instead just estimating a “cocycle”, which describes how a conditional distribution changes under causal actions (eg interventions). 🔥In various situations, directly targeting the cocycles simplifies the estimation problem and improves robustness to misspecification. 💪🏼This means more reliable, robust answers to causal “what if” questions in practice! Anyone interested can find out more in our paper (see the link in the comments)! Thank you to Stefan Feuerriegel for the invitation and for organising a great selection of speakers covering causal discovery and estimation problems! It’s always inspiring for the more theoretical end of my work to learn about the applied problems researchers in academia and industry are grappling with!

81

1 Comment

Check out this highly recommended course for those interested in optimizing their models: - Introduction to Quantization - Symmetric and Asymmetric Quantization Modes - Hands-on Implementation from Scratch using PyTorch - covering Per Tensor, Per Channel (Axis), and Per Group techniques - Weights Packing: Exploring 2 or 4 Bits Quantization Methods https://lnkd.in/gMAuWXHm

43

In this week's blog post, Data Science Fellow Sharon H. Green, PhD, MPH writes about how to create powerful visualizations for propensity score matching in R. If you're new to causal inference and trying to learn what tools are available to you, this is a great blog post to check out! Subscribe to our Medium for weekly blog posts on all topics data science, social science, and social justice. https://lnkd.in/epgPtduB

4

🏆 Meet MSDS Kaggle Competition Winners! 🏆 Congratulations to Eren Bardak, Karthik A., and Nithin Kumar from University of San Francisco Master of Science in Data Science Advanced ML course for acing the competition! In our latest podcast episode, they share their strategies and lessons learned, including how they navigated massive datasets and handled foreign language data. 🎧 Tune in now: https://lnkd.in/gkxQq2Pw #USFCA #MSDS #Podcast #DataScience #Cohort12 #MachineLearning #Kaggle #AI #TechTalk

7

1 Comment

Check out Tim Chartier's latest article, "Data Science Distance: When Close Seems Far," now available online and free to access in Scatterplot: The MAA Journal of Data Science. Dive into the complexities of distance metrics and how they impact data science. Scatterplot provides research and expository content at the intersection of mathematics, data science, and education. Explore this insightful read and stay updated with cutting-edge research in the field. 🔗 https://bit.ly/4c2b4VY

7

Our code for RAG evaluation is now publicly available at Amazon Science website: https://lnkd.in/g-HKUNZZ. This is the implementation of our evaluation mechanism performed by scoring the RAG on an automatically-generated synthetic exam composed of multiple choice questions based on the corpus of documents associated with the task. #OpenSource #OpenSourcing #AmazonScience #ICML2024 #Evaluation #LLMEvaluation

213

6 Comments

Follow along Jarom Hulet's clear and concise explainer to understand why homoscedastic errors are required for valid standard errors in linear regression, how to calculate heteroscedastic robust errors, and why the latter remove the need for the homoscedastic assumption.

19

A new paper titled "Which Humans?" by Mohammad Atari, Mona J. Xue, Peter S. Park, Damián E. Blasi, and Joseph Henrich unveils the overlooked psychological diversity in Large Language Models (LLMs). 🧠✨ The study highlights the prevalent "WEIRD" bias (Western, Educated, Industrialized, Rich, and Democratic) in LLM training data, aligning more with WEIRD populations than the global human population. 🔍 By comparing LLM outputs with the World Values Survey, the research finds significant cultural disparities. LLMs reflect WEIRD thinking styles, notably from nations like the US, Netherlands, Finland, Sweden, and Ireland, overshadowing non-WEIRD perspectives. The authors call for diversifying training data to truly reflect global human psychology, addressing both scientific and ethical challenges. 🌍📊 #ArtificialIntelligence #MachineLearning #PsychologicalDiversity #WEIRDBias #AIResearch #GlobalPsychology https://lnkd.in/gYeMSd7E

2

Sharing my latest article on the RAFT: Teach LLMs to be better at RAG Discover how RAFT bridges the gap between fine-tuning and RAG, enhancing model performance by integrating domain knowledge with powerful retrieval techniques. 🔍 Dive into: • The mechanics of RAFT • Benefits over traditional methods • Real-world applications using Meta Llama 2 Read the full article here: https://lnkd.in/g7ZQNE5s #AI #MachineLearning #DataScience #NaturalLanguageProcessing #AIInnovation #MetaLlama2 #RAFT #llm #RAG #genAI #LargeLanguageModels

9

#ITiCSE paper #3: 🔥 Compiler AND runtime errors predicting programming performance. In this paper we compared measures of student struggle based on both compiler and runtime errors. Both types of errors play role in predicting students' final grades.

1

🚀 Check out the latest Medium post by our student Tianyunxi Yin: "Quantization Fundamentals for Model Compression". Explore the essentials of quantization and learn how it helps in compressing machine learning models for better efficiency without compromising performance. Read more here: https://lnkd.in/gHbbR_J4 #USFMDS #DataScience #MachineLearning #ModelCompression #Quantization

15