Probability weight is a future choice weight based on probability calculation. I have set up a simple framework for probability weights: 1. The probability weight calculated based on expected values (related to space). The people who have won the lottery in history have all taken advantage of the positive expectations that occasionally appear in the lottery pool. So they seize the opportunity to buy desperately, the more they buy, the closer they are to the expected value under the law of large numbers. On the other hand, the absolute return is also greater. However, if faced with negative expectations, it's no use stubbornly kowtowing. Diligence is meaningless for negative expectations such as gambling and buying lottery tickets. 2. Probability weights based on Bayesian updates (time dependent). The quick trial and error in entrepreneurship aims to continuously optimize the probability of business models through Bayesian updates, until positive expected arbitrage opportunities are discovered. A powerful person will keep throwing dice to see what the dice say. This is the simulation of Monte Carlo, which cleverly makes mistakes within a radius that can converge. Not only do you learn from others, but you also dare to be the dice yourself. The Bayesian school believes that simulating uncertainty is the key to learning and utilizes Bayesian and Markov networks to work.【make money,Google,guessing competition,】