The competition has four tracks. The Gameplay and Learning tracks resemble traditional reinforcement learning competitions, the Level-Generation track focuses on the generation of entertaining game levels, and the Turing Test track focuses on humanlike game-playing behavior. We also outline some lessons learned from the competition and its future. The article is written by the four organizers of the competition. Within the CI/AI in games community, a series of competitions has grown up where competitors submit controllers for modified or reconstructed versions of existing computer games.

This article describes the winning entries in the 2000 American Association for Artificial Intelligence Mobile Robot Competition. The robots, developed by Swarthmore College, all used a modular hybrid architecture designed to enable reflexive responses to perceptual input. Within this architecture, the robots integrated visual sensing, speech synthesis and recognition, the display of an animated face, navigation, and interrobot communication. As any AI researcher knows, developing one robot for a real-world task such as serving hors d'oeuvres to a crowded room can be a significant undertaking. Our plan was to have a team of three agents participate in the hors d'oeuvres-serving contest and have one mobile robot attempt the Urban-Search-and-Rescue (USAR) event.

I wonder what will happen if I press this button? Algorithms armed with a sense of curiosity are teaching themselves to discover and solve problems they've never encountered before. Faced with level one of Super Mario Bros, a curiosity-driven AI learned how to explore, avoid pits, and dodge and kill enemies. This might not sound impressive – algorithms have been thrashing humans at video games for a few years now – but this AI's skills all were all learned thanks to an inbuilt desire to discover more about the game world. Conventional AI algorithms are taught through positive reinforcement.

Procedural content generation (PCG) for videogames relies on a commitment to the semantics of the game. Concepts such as enemies or solidity are required for the creation of levels for platformer games. As humans, we can instantly identify the underlying semantics of a game from brief snippets of game play video or from playing the game. Previous PCG systems have needed humans to identify the semantic properties of objects in the game, either implicitly or explicitly. We propose a system that can automatically learn the semantic properties of game objects by observation of events in the game via a causal learning framework. We apply this learning approach to play traces from the Super Mario Bros. series.

On Sunday, the Japanese prime minister turned up at the Olympics closing ceremonies to promote the 2020 Tokyo games dressed up as Mario, the eponymous hero of the popular video game series created in 1985. Who is Mario and how did he come by his global appeal? Japanese animation and game characters from Hello Kitty to Pac Man also made appearances in the closing ceremonies. But none of them may have the global reach of Super Mario, the game franchise that was a hit when Nintendo's video game system and Game Boy burst onto the scene in the 1980s. Prime Minister Shinzo Abe's appearance as Super Mario was a crowd-pleasing reminder of how much the game helped spur on the videogame revolution in the U.S. and globally.

On Sunday, the Japanese prime minister turned up at the Olympics closing ceremonies to promote the 2020 Tokyo games dressed up as Mario, the eponymous hero of the popular video game series created in 1985. Who is Mario and how did he come by his global appeal? Japanese Prime Minister Shinzo Abe appeared as the Nintendo game character Super Mario at the Olympics closing ceremonies to promote the 2020 Tokyo games. Mario Brothers is the best-selling video franchise of all time, with more than 490 million units sold since 1995, according to research firm NPD. There are more than 100 games, for various gaming systems, ranging from Donkey Kong to Super Mario Kart, in which Mario is the primary character, and many more in which he makes appearances.

The sacred Olympic flame can surely be trusted to stay lit in the care of Godzilla -- a famous Japanese character in the form of a giant, fire-breathing monster. There's literally a game called "Mario & Sonic at the Olympic Games," so when Shinzo Abe appeared as Mario, he left many fans wondering where exactly Sonic was. Hopefully Tokyo will take advantage of this beautiful opportunity and give Sonic the attention he deserves. If you think the real 2020 Olympics are gonna be cool wait till you see Mario and Sonic at the Tokyo games. I'm just saying Mario and Sonic at Tokyo 2020 SEGA, you have the Miku license I'M JUST SAYIN' If Tokyo 2020 isn't just a huge recreation of Mario and Sonic at the Olympic Games, I'm going to be thoroughly disappointed Can Mario please be the icon of the 2020 Olympics? And then Mario's face is literally the logo of Mario and Sonic 2020 games

In case you didn't notice, the Olympics is wrapping up in Rio, and sports fans now have an endure a whole three years and 11 months until the next one, which happens to be in Tokyo. Japan's Olympic committee decided to highlight that fact by transforming the country's Prime Minister into Mario (naturally), showcasing its distinctive skyline (swoon!), showing some of its athletes you've probably never heard of (sorry!) and a bunch video game and anime characters. It's all backed by a punchy jazz soundtrack, and is likely to give you goosebumps, although I wonder why the creative team wasn't able to book Pikachu in time.

In an epic teaser for the 2020 Tokyo Olympic Games, the prime minister of Japan arrived at the Rio Games through a green pipe dressed as Mario. If that doesn't get you excited for the upcoming Games, nothing will. In that one perfect moment, Shinzo Abe became everyone's hero. PM Abe appears at Rio closing ceremony dressed as Mario. I like the looks of the next Olympics.

Reinforcement learning describes how a learning agent can achieve optimal behaviour based on interactions with its environment and reward feedback. A limiting factor in reinforcement learning as employed in artificial intelligence is the need for an often prohibitively large number of environment samples before the agent reaches a desirable level of performance. Learning from demonstration is an approach that provides the agent with demonstrations by a supposed expert, from which it should derive suitable behaviour. Yet, one of the challenges of learning from demonstration is that no guarantees can be provided for the quality of the demonstrations, and thus the learned behavior. In this paper, we investigate the intersection of these two approaches, leveraging the theoretical guarantees provided by reinforcement learning, and using expert demonstrations to speed up this learning by biasing exploration through a process called reward shaping. This approach allows us to leverage human input without making an erroneous assumption regarding demonstration optimality. We show experimentally that this approach requires significantly fewer demonstrations, is more robust against suboptimality of demonstrations, and achieves much faster learning than the recently developed HAT algorithm.