Reinforcement Learning, Buffett, Poker, and the Wisdom of “Man Proposes, Heaven Disposes”

What do artificial intelligence, Warren Buffett, and poker have in common? They all remind us that success isn’t about luck or short-term outcomes, but about making consistently good decisions in the face of uncertainty.

In the 1960s, reinforcement learning — the branch of AI that learns by trial and error — hit a wall. The problem was known as credit assignment. When a system finally achieved a good outcome, it had no idea which earlier actions deserved credit. It’s like celebrating a successful product launch but not knowing whether it was the marketing campaign, the engineering team, or just dumb luck that made it work.

Then came a breakthrough: Temporal Difference (TD) Learning. Instead of waiting until the end to hand out rewards, TD learning gives credit along the way — rewarding good decisions as they happen, even before the final result is known. It taught machines to evaluate actions in context, not just outcomes in hindsight.

If that philosophy sounds familiar, it’s because Warren Buffett has been applying it in business for decades. Buffett’s approach to management compensation isn’t about rewarding outcomes, which can be distorted by market cycles or macroeconomic randomness, but about rewarding sound decisions made within one’s control. A manager who makes disciplined, rational choices should be celebrated, even if short-term results don’t cooperate.

The same principle governs poker strategy. The best players know that even perfect decisions can lose in the short run. Luck can deal you a bad hand or a nasty beat. But a great poker player focuses on process over outcome — making decisions with the best probabilities and information available. Over the long run, those good decisions compound into consistent wins.

This timeless truth is captured perfectly by an old saying:
"Man proposes, Heaven disposes."
We can control the quality of our plans and decisions, but the final outcome often lies in forces beyond our influence.

Interestingly, Richard Sutton, one of the fathers of temporal difference learning, didn’t dream this up in a vacuum. His inspiration came from observing animals. In nature, learning is messy, iterative, and incremental. Evolution doesn’t hand out rewards at the finish line — it shapes behavior over millions of small feedback loops. Survival itself is one giant reinforcement learning experiment.

And maybe that’s the most profound insight of all: real-world intelligence evolves not by chasing results, but by refining decisions.

Whether you’re building AI, managing a company, playing poker, or navigating life, the lesson is the same — focus on the decisions you can control, reward the process, and trust that the results will follow in due time.

评论

发表评论

此博客中的热门博文

Silicon vs. Carbon: A Tale of Two Intelligences

Silicon Intelligence vs. Carbon Intelligence: A Tale of Two Energies “Artificial Intelligence” is a misnomer, if you ask me. I’d rather call it Silicon Intelligence . After all, it runs on silicon chips (at least for now) and is powered by electricity. By contrast, human intelligence is carbon-based—built from DNA and fueled by bioelectricity. Ultimately, all that energy—whether from a GPU cluster or a human brain—traces back to the sun : The past sun powers us via coal, oil, and gas. The present sun drives wind turbines, solar panels, and photosynthesis (plant food included). Preset Parameters and the Art of Learning When a baby is born, much of their foundational “mental model” is wired by genetics. As Charlie Munger famously quipped, “They’re like machines with 80% of the parameters preset.” But humans are also learning machines . We don’t recompile—we reflect. And if we’re lucky, we wake up “a little less stupid” each day. Large language models improve similarly—via fine-...
阅读全文

The Ghost in the Machine: How Human Cognitive Biases Shape the Alignment, Context, and Tuning of Large Language Models

 The entire discipline of guiding LLM outputs can be reframed through the lens of "choice architecture," a concept from behavioral economics introduced by Richard Thaler and Cass Sunstein. Choice architecture is the design of the environment in which people make decisions; it posits that the way choices are presented—the "architecture"—inevitably influences the outcome, whether intentionally or not. A choice architect, therefore, is anyone who organizes the context in which people make decisions. In the realm of LLMs, prompt engineers and the designers of Retrieval-Augmented Generation (RAG) systems are the new choice architects. They are not merely instructing a machine; they are designing a decision-making environment for an artificial intelligence. This section argues that this process is profoundly shaped by the same heuristics and biases that affect human decision-making, turning context engineering into a practice of creating flawed choice architectures.
阅读全文

For Whom the Bot Toils: Navigating the Great Inequality of the Gen AI Era

 We stand at the precipice of a revolution, one powered not by steam or silicon, but by syntax and servers. The rise of Generative AI has been heralded as the dawn of democratized intelligence, a future where every individual has a brilliant co-pilot in their pocket. Yet, as we watch this new world take shape, a rather inconvenient and increasingly stark reality is emerging: the algorithm giveth, and the algorithm taketh away—on a scale that would make John D. Rockefeller blush. The economic landscape is beginning to look less like a level playing field and more like a cartoon where one character has a rocket ship and everyone else has a pogo stick. On one hand, we see super-intelligence labs at companies like Meta offering compensation packages to top AI researchers that resemble the GDP of a small island nation. These are the new high priests of progress, crafting the digital deities that will define our future. In parallel, tech titans like Microsoft and Nvidia are soaring towar...
阅读全文