Reinforcement learning is a paradigm through which an agent interacts with its environment by trying out different actions at different states and observing the outcome. Each of these interactions can change the state of the environment, and can also provide rewards to the agent. The goal of the agent is to learn the value of performing each action on each state. By value, we mean the biggest amount of rewards that is possible for the agent to obtain after performing that action in that state. If the agent achieves this goal, it can then act optimally on its environment by choosing, at every state, the action that has the biggest value.

All the sessions from Transform 2021 are available on-demand now. As consumers, we've widely welcomed artificial intelligence and machine learning into our daily lives. "Smart" speakers, facial recognition on our phones, targeted ads we love to hate -- these are just some of the AI-powered technologies all around us. But inside companies, where AI holds virtually incalculable benefit in a range of use cases -- such as hyper-efficient and productive IT, supply chain automation, and increasingly intelligent cybersecurity ecosystems -- the status of adoption is more of a mixed bag. In a recent survey of 700 IT pros across the globe, a whopping 95% said they believe their companies would benefit from embedding AI into daily operations, products, and services, and 88% want to use AI as much as possible. In the trenches, IT staffers see AI as a way to help them do their jobs faster and better, and they're gravitating toward it as naturally as consumers have gratitated toward smart speakers at home.

In Dynamic Programming (DP) we have seen that in order to compute the value function on each state, we need to know the transition matrix as well as the reward system. But this is not always a realistic condition. Probably it is possible to have such thing in some board games, but in video games and real life problems like self-driving car there is no way to know these information before hand. If you recall the formula of the State-Value function from "Math Behind Reinforcement Learning" article: It is not possible to compute the V(s) because p(s',r s,a) is now unknown to us. Always keep in mind that our goal is to find the policy that maximizes the reward for an agent.

In my last post, I wrote about Artificial Intelligence (AI). When I last wrote about AI, I focused on the technological side: what is a part of an AI system and what isn't. However, there is another question which might be more important; what are we doing with AI? Part of my job is to help investors looking at AI companies with their due diligence. I have discussions with them about companies they might want to invest in. Through this process, I have observed how every company pitch is full of content on how they are using AI to solve a business problem.

I have written on Artificial Intelligence (AI) before. Back then I focused on the technology side of it: what is part of an AI system and what isn't. But there is another question which might be even more important. What are we DOING with AI? Part of my job is to help investors with their due diligence. I discuss companies with them in which they might want to invest.