hidden Markov model
A probabilistic statistical model that assumes a system has hidden states and uses observed events to infer the likelihood of those hidden states, commonly used for sequence prediction
As defined in The Story of Jim Simons: The World’s Most Successful Investor
Other definitions
A statistical model that infers unobserved states from sequences of observable events, such as user behavior, using transition and emission probabilities.
Making Decisions with Data
Where it comes up · 2
- The Story of Jim Simons: The World’s Most Successful Investor2 claims
- 363 | Chandra Sripada on How LLMs and Humans are Cognitive Cousins1 claim
Most important claims · 3 of 3Search all claims for “hidden Markov model” →
- In 1968, Jim Simons published an internal IDA paper proposing that a quantitative firm using hidden Markov models could generate 50% annual returns, which led to his dismissal after he publicly stated he would work 0% for IDA until the Vietnam War ended.The Story of Jim Simons: The World’s Most Successful Investor
- When a prompt supplies Einstein is German and Gandhi is Indian before asking Marie Curie is..., the model infers the task pattern and answers French, and studies using mixtures of hidden Markov models show this resembles implicit Bayesian inference over candidate tasks.363 | Chandra Sripada on How LLMs and Humans are Cognitive Cousins
- A hidden Markov model is a probabilistic model that identifies hidden patterns based on observed events, such as using humidity to help predict future weather patterns beyond simple observable weather indicators.The Story of Jim Simons: The World’s Most Successful Investor