Jiun Kim

This months instalment of my Causal AI series is up on Towards Data Science, this time exploring optimising non-linear treatment effects in Pricing & Promotions (or Marketing): https://lnkd.in/etYraNxt

101

2 Comments

2024 is the year of AI value, and time for companies to get out of PoC purgatory. “I really want to start driving use of AI at scale,” Luke G., head of analytics and AI at TD told The Wall Street Journal. “I think now is the time.” https://gag.gl/w2TUIR In his recent RevX keynote, Domino Data Lab's Nick Elprin detailed how companies do this today. Watch and learn how to accelerate AI impact: https://gag.gl/oN6hJo #AI #datascience #GenerativeAI

1

In my latest Towards Data Science article, we shift our focus to understanding how multi-collinearity can damage the causal inferences you make, particularly in marketing mix modelling. Should be a good read for those of you interested in MMM: https://lnkd.in/eVErcCit

99

4 Comments

I am glad to share my views on Anthropic's latest paper: Scaling Monosemanticity. Thank Towards Data Science for publishing it! #monosemanticity #interpretability #LLM #GenAI

23

For anyone interested in a quick insight from “Strawberry” or o1-preview, I have found that it is the most capable LLM I have worked with to help construct a virtual environment and navigate dependency issues (everyone’s favorite part of a build). I do recommend a new chat frequently because it does, seemingly, get a bit lazy. Currently it is helping me train an instance of Apple’s OpenELM 3B where all of the files are downloaded on device. If you have questions please reach out. #openelm #chatgpt #apple

3

2 Comments

https://lnkd.in/esz3v_U2 Inspired by John Forstmeier to `build in the open`, and to see if my $100 webcam works, going to start doing some livestreaming, working on this side project I've been meaning to prioritize. The idea is I want to build out better tooling for ML so more people can jump into the field and mess around. Project: https://lnkd.in/ey_Bk7m2 Gonna start by just messing around with some Fine Tuning Llama, just because I haven't done that in a minute. Come hang out or watch it when it's finished.

6

In the wake of genAI hype, developers of all stripes are striving to design and implement application architectures that make use of the advantage of #LLMs and appropriately manage their disadvantages. 2024 is year 1 for optimizing genAI applications, and today you've probably heard of Retrieval Augmented Generation (RAG) as one of the common approaches. But what is #RAG in a business context? How can it be applied to enterprise data and applications? And are there other approaches to customizing the behavior of foundation models for language like #LLMs? To get more information on the first question, check out Rowan Curran’s recently published pair of Forrester blogs on RAG and what comes next in #genAI-infused application: #RAG is all the Rage (Part 1): https://lnkd.in/eTHwdJCU The Machine is Getting More Complex (Part 2): https://lnkd.in/eWWVzDve If you'd like to learn more about the latter - adapting LLMs to your enterprise - come to my session in September at #Forrester's Technology and Innovation forum: https://lnkd.in/eWnSMNWX

6

Episode 6 of the Data Science Rabbit Podcast is finally out! 🔘 Direct Preference Optimization for training Generative AI models 🔘 Tips on data visualization 🔘 Revisiting a "graduation speech" about AI that I wrote last year (inspired by the so-called "Wear Sunscreen" essay by Mary Schmich) Available at all the major podcast distributors, including Spotify: https://lnkd.in/edYTw988 #DataScience #Podcast #AI #ArtificialIntelligence

5

2 Comments

In my latest article on Towards Data Science we turn our attention to powering experiments using CUPED and double machine learning: https://lnkd.in/etcGpUU7 Should be an interesting read for anyone wanting to learn more about experimentation and AB testing. Let me know what you think 🙂

41

3 Comments

The next part of my Causal AI series is up on Towards Data Science, this time exploring how Linear Programming can be used to optimise treatment strategies: https://lnkd.in/eb2ErFgi

46

📊"Not Your College Stats Course: Engaging Stakeholders Through Data Science" 🎥 In the industry, data scientists must collaborate with colleagues across various roles. This talk from #NYRConf dives into how to effectively communicate results to non-technical teams and build stronger relationships with stakeholders. Learn techniques to break down complex analyses and align with stakeholder needs. Watch now! 📈💡 #DataScience #StakeholderEngagement #RStats 🔗 https://lnkd.in/egjFT8fv 

1

Those interested in interoperability, including laboratory interoperability should comment through August 23rd on the expanded Data Usability Implementation Guide. The guide focuses on health data exchange across EHR, HIE, Public Health, and laboratories, leveraging published standards to lift everyone up together and help retain vital information in exchanges for downstream users.

6

2 Comments

Inspired by Ken Flerlage's 2020 Tableau dashboard on different ways to visualize election data, I built something similar using state-level 2024 presidential election results in Flourish. Check out all of the 15 different approaches here: https://lnkd.in/eCWvqG4z

18

Sign up now for our Introduction to Causal Inference course, where Mulusew J. Gerbaba Associate Research Scientist – Impact Evaluation at APHRC will dive into the methods of causal inference using observational data. You'll explore topics like potential outcomes, DAGs and causal graphs, target trial emulation, matching techniques, managing confounding and bias, causal structure learning, and quasi-experimental methods. Don't miss this opportunity to acquaint yourself with modern causal modeling methodology. Register today! https://lnkd.in/dcMnHVzX #MentalHealthDataPrizeAfrica #MentalHealthAwareness #DataForChange

10

Exciting to see Eric Korman featured in Data Science Spotlight, sharing valuable perspectives on techniques that data scientists can use to ensure reproducibility in data analysis workflows. In a field where reliable and repeatable results are paramount, Eric shares a small bit of knowledge on the methodologies and best practices that empower data teams to deliver consistent, trustworthy insights. Read his response here: https://bit.ly/3NTnghu #DataScience #Reproducibility #DataIntegrity Featured

14

Data-enabled missions—those that have the technology, talent, policies, and processes in place to leverage the full spectrum of relevant data toward achieving positive mission outcomes—will be critical to advancing national prosperity in the face of today’s complex, global challenges. Where is your agency on its #data enablement journey? Learn More: https://lnkd.in/gE2jPN2J https://lnkd.in/g_gjxKeA

1

Simon Willison 's term “imitation intelligence” is such a great way to think about LLMs and generative AI. If we remember that models *appear* intelligent through token prediction and aren’t actually “smart” (yet?), we'll be better users. It's a useful framing for both their capabilities and limits. His #PyConUS presentation is an interesting watch/read/listen, whether or not you use Python: https://lnkd.in/e5U7EM3G #GenAI #LLMs #AI #ArtificialIntelligence 

5

AI Agents vs. Agentic AI: Are Our Digital Assistants Getting a Mind of Their Own? Just shared a new blog exploring where AI is headed—from the familiar task-based assistants we know to a new breed of proactive, agentic AI. We’re talking about AI that doesn’t just wait for us to ask for help but gets what we need done before we even ask. Give it a read and let me know what you think :) #AI #AgenticAI #DigitalFuture #AIThoughts #Innovation

7

1 Comment