Tags » Reinforcement Learning

Tabula Rasa Reinforcement Learning: self learning AI

Here is an interesting article on self learning AI models:

The game of chess is the most widely-studied domain in the history of artificial intelligence. The strongest programs are based on a combination of sophisticated search techniques, domain-specific adaptations, and handcrafted evaluation functions that have been refined by human experts over several decades.

97 more words
Data Science

Playing Frozenlake with genetic algorithms

Our topic for today will be using Random Policy and enhance it with genetic/ evolutionary algorithms to score in different versions of FrozenLake.

About FrozenLake, OpenAI gym:

560 more words

Shortest path finder in Processing

Simon created a wonderful project in Processing – a path finder that looks for the shortest path to reach the green cell, avoiding the obstacles. Every time the path finder fails, it tries again. 79 more words


The beauty of greed

December: the month of shopping and exquisite cooking. Time for a desire theme. Watching two documentaries on capitalism (Doughnut Economics and Saving Capitalism), I asked myself: why do people who own so much already stay so greedy? 423 more words


Standard Bandit: Convergence

In this article we explore convergence of bandit algorithms. By convergence we mean optimal action (arm) is learned and/or applied. Specifically convergence can be interpreted differently below… 241 more words

Unknown Boundary Go?

Thinking about the differences between what currently passes by “machine intelligence” and human intelligence, I came up with a variant of the game of Go that I have not seen mentioned anywhere. 245 more words


Standard Bandit: Computational Cost

Computational cost is important for very large action space, or when the bandit is incorporated as part of Monte Carlo sampling. For example, in an e-commerce environment, a company needs to select one product out of millions (if not billions) of candidates within a fraction of a second. 216 more words