Access and Feeds

Artificial Intelligence: AutoML: Self-Evolving Algorithms

By Dick Weisinger

Artificial Intelligence (AI) algorithms have been criticized because their developers purposely, but more likely, inadvertently, embed biases and prejudices into the design and workings of the AI.

A report by AI Now finds that built-in biases “affect how AI companies work, what products get built, who they are designed to serve, and who benefits from their development.”

Is there a solution? Some say that algorithms designed not by humans but by code could eliminate or reduce built-in prejudices. Some think that a technique known as AutoML might be able to do that. AutoML is machine learning that uses one generation of AI algorithms to design and write a new generation of AI code.

But Research by Google concluded though that “human-designed components bias the search results in favor of human-designed algorithms, possibly reducing the innovation potential of AutoML. Innovation is also limited by having fewer options: you cannot discover what you cannot search for.”

The hope is that AutoML-discovered algorithms may be able to get past biases, and at that point, AI might be able to go much further than we might imagine.

Haran Jackson, the chief technology officer (CTO) at Techspert, said that “there is a sense that among many members of the community that the most impressive feats of artificial intelligence will only be achieved with the invention of new algorithms that are fundamentally different to those that we as a species have so far devised.”

Digg This
Reddit This
Stumble Now!
Buzz This
Vote on DZone
Share on Facebook
Bookmark this on Delicious
Kick It on
Shout it
Share on LinkedIn
Bookmark this on Technorati
Post on Twitter
Google Buzz (aka. Google Reader)

Leave a Reply

Your email address will not be published. Required fields are marked *


sixteen − eleven =