Win Vector LLC
Data science advising, consulting, and training
- Website
- win-vector.com · all feeds from this site
- Feed URL
https://win-vector.com/feed/
Read it in your FeedListing reader
Or add to: Feedly Inoreader Raw feed
Latest posts
-
“Boy Born on Tuesday” A Thornier Problem Than The Monte Hall Problem?
The standard “treacherous” probability problem is the famous Monty Hall problem. Some of the issue is translating a word problem into math terms, and a lot of it is unfamiliarity with probabilistic reasoning (most majors don’t spend a lot of time on this). I think I have found…
-
Speaking at the Bayesian Interest Group November 2026 Virtual Meeting
I’ll be speaking at the Bayesian Interest Group November 2026 Virtual Meeting. The topic is using Bayesian models to interactively pre-estimate the value of applying an expensive inspection or procedure to a large population. For example: estimating the value of using an LLM…
-
How to Critically Read LLM Texts [Video]
I’d like to share a new talk on how to critically read texts produced by LLMs. I supply a mental model to try and help prime the reader for some of the things to look for in such a text, and to perhaps be a bit less surprised by what […]
-
A Simplified Mental Model of LLMs
Introduction As of now (late 2026) LLM (large language model) technology providers, users, and work products flood the public commons. It therefore makes sense to have even a primitive mechanistic mental model of these technologies. You are forced to have an opinion. Without a…
-
Expected Log-Return as a Risk Control
The question Here is a betting/investment puzzle to help introduce some tools for reasoning about risk. You are offered a bet or risky investment that returns or pays-off X dollars for every dollar laid in, where X is a random variable distributed log-normally with parameters μ…
-
Choose by Probability, not by Expectation
A lot of financial thinking and intuition attempts to phrase decisions and choices in terms of claimed returns or in terms of expected values. For example: An ultra greedy gambler may pick the highest claimed winning payoff, independent of win or success probabilities. A greedy…
-
LLMs Will Benefit from Scratch Workspaces
Current (mid-2026) LLMs are quite powerful. However, (prior to addition of agent harnesses and loops) LLMs vendors don’t claim to include an effective scratch workspace. Current LLM designs, to the extent we know what they are, remain extensions of the attention/transformer…
-
Maybe Jacobian Conjecture Counterexamples are not Rare
The (now falsified) Jacobian conjecture was a well known conjecture expressing the hope that a family of functions had nice properties. In July 2026 Levent Alpöge exhibited a simple counter-example that he said was found by Claude Fable 5. Fields medalist Terence Tao wrote on…
-
Is it the box or the “yes” that does not fit?
There is a famous puzzle where a boy takes a 5 foot fishing rod on a bus that doesn’t allow objects taller than 4 feet on by packing it in a 3 by 3 by 3 box (which has a long diagonal over 5 feet). The idea being: such a […]
-
Why You Are Not Getting Full Value From Your Analytics
We have a short pitch called “Why You Are Not Getting Full Value From Your Analytics” that we would like to share (and hope some of you will re-share in your network). My group, Win Vector LLC, has some neat methods to extract more value out of business analytics called […]