Students Create Doom AI Which Learns Visually and Kills Humans in Deathmatch | Hacked: Hacking Finance
user

Artificial Intelligence

Students Create Doom AI Which Learns Visually and Kills Humans in Deathmatch

Posted on .

Students Create Doom AI Which Learns Visually and Kills Humans in Deathmatch

Introduction

This article was posted on Friday, 14:49, UTC.

Two students from Carnegie Mellon University recently placed second in an artificial intelligence competition for their submission of a program that was able to learn the game Doom the same way humans do – by playing.

// -- Discuss and ask questions in our community on Workplace. Don't have an account? Send Jonas Borchgrevink an email -- //

Whereas most such programs have traditionally learned by following a set of instructions and subsets therein if this, do that), the program learns based on what’s happening on the screen. In one round, the AI even learned to duck, a favored tactic of human players.

Guillaume Lample and Devendra Singh Chaplot spent four months creating a program that utilized many different facets of AI technology. The first problem they solved was that of “rewarding” the AI. The AI must have a framework to learn what is good and bad behavior. In their paper on the subject, the students note that if the reward is based on score alone, the learning is slow because that reward does not appear until the end of the round.

// -- Become a yearly Platinum Member and save 69 USD and get access to our secret group on Workplace. Click here to change your current membership -- //

Instead, they introduced a concept learned about from a 2003 paper by Baidu Research Chief Scientist Andrew Ng called “reward shaping.” Reward shaping allows their bot to gain intermediate rewards that build up to a complete reward. Their reward table looked like this: positive reward for picking up objects, negative reward for losing health, and negative reward for shooting/loosing ammunition.

We used different rewards for the navigation network. Since it evolves on a map without enemies and its goal is just to gather items, we simply give it a positive reward when it picks up an item, and a negative reward when it’s walking on lava.

We also found it very helpful to give the network a small positive reward proportional to the distance it travelled [sic] since the last step. That way, the agent is faster to explore the map, and avoids turning in circles.

Other important factors go into the development of the bot, including map learning and a concept called “frame skip.”

Having placed second in a competition that sees a lot of heavy hitters entering it, the two are moving onto more advanced games like Quake. The development of advanced artificial intelligence in gaming could prove to be a very important aspect of the future of gaming as a whole. From a developer’s perspective, more advanced AI means that people will have more fun and be more challenged when playing first-person shooter games in single-player mode. From a player’s perspective, the prospect of cheating arises.

As more and more games see the introduction of paid-to-play models and lucrative competitions, game developers and publishers will have to become hyper-aware of how to differentiate between a bot and a human player. Both Doom and Quake are very old games, but the concept of the first person shooter has only evolved so much since their introduction, and the technology behind these bots isn’t so many leaps in logic behind them.

Featured image from Flickr.

Important: Never invest money you can't afford to lose. Always do your own research and due diligence before placing a trade. Read our Terms & Conditions here.



Feedback or Requests?

P. H. Madore

P. H. Madore

http://phm.link

P. H. Madore has covered the cryptocurrency beat over the course of hundreds of articles for Hacked's sister site, CryptoCoinsNews, as well as some of her competitors. He is a major contributing developer to the Woodcoin project, and has made technical contributions on a number of other cryptocurrency projects. In spare time, he recently began a more personalized, weekly newsletter at http://ico.phm.link

There are no comments.

View Comments (0) ...
Navigation
The team:
Dmitriy Lavrov
Analyst
Dmitriy Lavrov is a professional trader, technical analyst and money manager with 10 years of trading experience. He covers Forex, Commodities and Cryptocurrencies. He is among the top 10 most Read More
Jonas Borchgrevink
Founder
Jonas Borchgrevink is the founder of Hacked.com and CryptoCoinsNews.com. He is a serial entrepreneur, trader and investor. He shares his own personal journey on Hacked.com. // -- Discuss and ask Read More
Mate Csar
Analyst
Trader and financial analyst, with 10 years of experience in the field. An expert in technical analysis and risk management, but also an avid practitioner of value investment and passive Read More
Mati Greenspan
Analyst
Senior Market Analyst at Etoro.com. // -- Discuss and ask questions in our community on Workplace. Don't have an account? Send Jonas Borchgrevink an email -- // Important: Never invest Read More
Rakesh Upadhyay
Analyst
Rakesh Upadhyay is a Technical Analyst and Portfolio Consultant for The Summit Group. He has more than a decade of experience as a private trader. His philosophy is to use Read More
Pamela Meropiali
Account Manager
Pamela Meropiali is responsible for users on Hacked.com. // -- Discuss and ask questions in our community on Workplace. Don't have an account? Send Jonas Borchgrevink an email -- // Read More
Joseph Young
Journalist
Joseph Young is a finance and tech journalist & analyst based in Hong Kong. He has worked with leading media and news agencies in the technology and finance industries, offering Read More
Decades ago Intel Co-Founder Gordon Moore observed that the density,…