Every human being in every society learns VERY early that current decision/action has future consequences both intended or unintended. This comes first from parental teaching/training or through enforcement of societal law and/or traditions. We are taught that “honesty is often the best policy”, “society will punish bad behaviors for its common good”, et al. Although what the consequences will be are uncertain and varies with person and circumstance (e.g., can you get away with murder?). The truism of the title of this article is universal. However, beginning in the second half of the 20th century to now, theorists begin to mathematically formalize this notion of a basic stochastic decision-theoretic problem. This became known variously as Deep Learning in Neural Networks, Backward Induction in Dynamic Programming, TPBVP in Optimal Control, and other names (for example Pontraigin’s Maximum Principle if you are from Russia) depending on the field of origin and/or the naming person. I have forgotten but one scholar even credited me with the invention of this idea which I do not deserve. Basically everyone is trying to reflect future consequences to the present when decision, optimal in some sense, if possible have to be made. All decisions become a static math problem which is far easier, but often still computationally impossible, to handle/solve. Adoption of this approach permeates all kinds engineering problems including the most famous - landing and returning three astronauts on the moon in 1969 safely. It is no exaggeration that many of us including myself earned our lifelong living because of this. It is fundamental to human civilization and basic to continued human existence. OK. Now you know as much about this idea as I do. But don’t accuse me of having an easy life. Converting theoretical idea to practical real life success still takes hard work on the part of many people at different levels of endeavor. You have to compete world wide not just locally with many people who are equally if not smarter than you. Being lucky and prepared are often necessary but far from sufficient conditions.

转载本文请联系原作者获取授权，同时请注明本文来自何毓琦科学网博客。
链接地址：https://blog.sciencenet.cn/blog-1565-1525807.html

上一篇：THE COMPLEX BUSINESS OF DYING