Recsys veteran: most ML problems are really recommendation problems
vboykis · x · 2026-09-08
vboykis, a practitioner who has been working continuously in recommender systems, argues she is increasingly convinced that most ML problems are fundamentally recsys problems. The observation points to how ranking, personalization, and user-behavior modeling generalize across many ML subfields—a useful lens for re-examining one's own modeling tasks.
More from AGI Musings
- OpenAI Chief Scientist Urges 'Extreme Caution' With Pace of AI Development — pstAsiatech · 2026-09-08
- AI-generated CAD/PCB is in its 'slop that gets clicks' stage, say devs — sebkrier · 2026-09-08
- Dan Faggella weighs in on Hendrycks' posthuman turn: avoid collapse first, then pick grand futures — danfaggella · 2026-09-08
- Maliciously trained AI agents will overshoot operator intent, blurring misuse and misalignment — pstAsiatech · 2026-09-08
- Gaming subreddit gives up anti-AI moderation fight, sparking skill-vs-AI debate — NathanpmYoung · 2026-09-08
- Gary Marcus says LLMs still haven't produced new linguistic generalizations, echoing Chomsky — GaryMarcus · 2026-09-08