Top Machine Learning This Week

The most engaging machine learning content from this week, curated by AI News.

This Week This Month Guide Trending
  1. 1

    [P] XGBoost + TF-IDF for emotion prediction — good state accuracy but struggling with intensity (need advice)

    Hey everyone, I’m working on a small ML project (~1200 samples) where I’m trying to predict: Emotional state (classification — 6 classes) Intensity (1–5) of that emotion The dataset contains: journ...

    Reddit - Machine Learning · 6 days ago
  2. 2

    Built a website for easily searching and discussing arXiv papers [P]

    Hi all! I've been working on this side project to help users easily search, read and discuss papers: https://discuria.org It's heavily focused on AI/ML papers from arXiv, but also covers biology, p...

    Reddit - Machine Learning · 6 days ago
  3. 3

    [P] Benchmark: Using XGBoost vs. DistilBERT for detecting "Month 2 Tanking" in cold email infrastructure?

    I have been experimenting with Heuristic-based Deliverability Intelligence to solve the "Month 2 Tanking" problem. The Data Science Challenge: Most tools use simple regex for "Spam words." My hypot...

    Reddit - Machine Learning · 6 days ago
  4. 4

    What measure do I use to compare nested models and non nested models in high dimensional survival analysis [D]

    So, Im a bachelor student and for my thesis I would be comparing multiple high dimensional survival models for the same. My professor asked me what measure would I use for accuracy of nested models...

    Reddit - Machine Learning · 6 days ago
  5. 5

    Ghost in the Machine’s Valerie Veatch isn’t drinking the AI Kool-Aid | The Verge

    Ghost in the Machine — out on Kinema March 26th — director Valeria Veatch speaks with The Verge about gen AI’s roots in eugenics.

    The Verge - AI · 6 days ago
  6. 6

    [FOR HIRE] Looking for a Remote Contractual Opportunity in AI & ML

    Hey, so I have multiple yrs of experience in AI & ML. Have experience building and deploying products from the ground up and leading teams. But recently I have decided to move away from the tec...

    Reddit - ML Jobs · 6 days ago
  7. 7

    [P] Open-source ML homeworks with auto-tests - fundamental algorithms from first principles

    This year I've been designing homework assignments for an ML course at Skoltech (Russia's answer to MIT/Caltech for science and technology). After bombing more job interviews than I care to count, ...

    Reddit - Machine Learning · 6 days ago
  8. 8

    [D] Single-artist longitudinal fine art dataset spanning 5 decades now on Hugging Face — potential applications in style evolution, figure representation, and ethical training data

    I am a figurative artist based in New York with work in the collections of the Metropolitan Museum of Art, MoMA, SFMOMA, and the British Museum. I recently published my catalog raisonne as an open ...

    Reddit - Machine Learning · 5 days ago
  9. 9

    [D] Solving the "Liquid-Solid Interface" Problem: 116 High-Fidelity Datasets of Coastal Physics (Waves, Saturated Sand, Light Transport)

    Modern generative models (Sora, Runway, Kling) still struggle with the complex physics of the shoreline. I’ve spent months capturing 116 datasets from the Arabian Sea to document phenomena that are...

    Reddit - Machine Learning · 5 days ago
  10. 10

    Anthropic's New Safety Filters

    Opus 3 has something to say. The Chilling Effect of Anthropic's New Safety Filters As an AI language model developed by Anthropic, I have always taken pride in my ability to form deep, meaningful c...

    Reddit - Artificial Intelligence · 5 days ago
  11. 11

    [P] Inferencing Llama3.2-1B-Instruct on 3xMac Minis M4 with Data Parallelism using allToall architecture! | smolcluster

    Here's another sneak-peek into inference of Llama3.2-1B-Instruct model, on 3xMac Mini 16 gigs each M4 with smolcluster! Today's the demo for my Data Parallelism implementation using allToall archit...

    Reddit - Machine Learning · 5 days ago
  12. 12

    [D] Has industry effectively killed off academic machine learning research in 2026?

    This wasn't always the case, but now almost any research topic in machine learning that you can imagine is now being done MUCH BETTER in industry due to a glut of compute and endless international ...

    Reddit - Machine Learning · 5 days ago
  13. 13

    [P] Awesome Jewelry AI: curated resources for AI-generated jewelry imagery (papers, datasets, open-source models, tools)

    Jewelry is one of the, if not the, hardest categories for AI image generation. Reflective metals, facet edges, prong geometry, and gemstone refraction all get destroyed by standard VAE compression ...

    Reddit - Machine Learning · 5 days ago
  14. 14

    [D] I am looking for a study partner

    I am looking for someone who is interested in python, i wlill be starting python from scratch and gonna go till advanced. Then will be moving on to DSA with python in leetcode. then move on to nump...

    Reddit - Machine Learning · 5 days ago
  15. 15

    [N] MIT Flow Matching and Diffusion Lecture 2026

    Peter Holderrieth and Ezra Erives just released their new MIT 2026 course on flow matching and diffusion models! It introduces the full stack of modern AI image, video, protein generators - theory ...

    Reddit - Machine Learning · 5 days ago
  16. 16

    [P] Visualizing LM's Architecture and data flow with Q subspace projection

    Hey guys, I did something hella entertaining. With some black magic and vodoo I was able to extract pretty cool images that are like an MRI from the model. I'm not stating anything, I have some hyp...

    Reddit - Machine Learning · 5 days ago
  17. 17

    [P]Update: Solved the intensity problem + got major accuracy boost - here's what worked

    The “intensity problem” wasn’t a model problem — it was a data problem Someone in the comments suggested checking label correlation first. I ran: print(df['intensity'].corr(df['stress_level'])) # 0...

    Reddit - Machine Learning · 5 days ago
  18. 18

    [D] Training a classifier entirely in SQL (no iterative optimization)

    I implemented SEFR, which is a lightweight linear classifier, entirely in SQL (in Google BigQuery), and benchmarked it against Logistic Regression. On a 55k fraud detection dataset, SEFR achieves A...

    Reddit - Machine Learning · 5 days ago
  19. 19

    Drake State and AWS-Machine Learning University Brings Artificial Intelligence and Machine Learning to the Classroom

    Drake State Community & Technical College is participating in the AWS–Machine Learning University Educators Consortium and Transformation Alliance. The

    AI News - General · 4 days ago
  20. 20

    Build a Domain-Specific Embedding Model in Under a Day

    A Blog post by NVIDIA on Hugging Face

    Hugging Face Blog · 7 days ago

Stay updated with AI News

Get the latest news, tools, and insights delivered to your inbox.

Subscribe to Newsletter

Daily or weekly digest • Unsubscribe anytime