Happy Nice Mall

Less than two years after Google fired two researchers who criticized the biases built into AI systems, the company fired a researcher who questioned an article it published about the capabilities of a special type of AI used in making computer chips.

The researcher, Satrajit Chatterjee, led a team of scientists to challenge famous scientists. research paperPublished last year in the journal Nature, which says that computers can design certain parts of a computer chip faster and better than humans.

43-year-old Dr. Chatterjee was fired in March, shortly after Google told his team it would not publish an article rebutting some of the claims made in Nature. To be important. In a written statement, Google said that Dr. He confirmed that Chatterjee was “terminated for cause”.

Google, Dr. He declined to elaborate on Chatterjee’s dismissal, but defended his reluctance to publish the research and assessment that he was critical of.

“We’ve thoroughly reviewed the original Nature document and stand by the peer-reviewed results,” Zoubin Ghahramani, vice president of Google Research, said in a written statement. “We also rigorously investigated the technical claims of a later application, and it did not meet our publication standards.”

Dr. Chatterjee’s dismissal was the latest example of the conflict in and around Google Brain, an AI research group considered key to the company’s future. After spending billions of dollars recruiting top researchers and creating new types of computer automation, Google has struggled with a wide array of complaints about how it creates, uses, and portrays these technologies.

The tension among Google’s AI researchers reflects much larger struggles in the tech industry, which is faced with numerous questions about new AI technologies and the difficult social issues that confuse those technologies and the people who create them.

The latest dispute also follows the familiar allegations of dismissals and duels among Google’s AI researchers; This is a growing concern for a company that is betting its future on placing AI in everything it does. Sundar Pichai, CEO of Google’s parent company Alphabet, compared artificial intelligence to the advent of electricity or fire and described it as one of humanity’s most important endeavors.

Google Brain started as a side project more than a decade ago when a group of researchers created a system that learns to recognize cats in YouTube videos. Google executives were so fascinated by the possibility that machines could learn skills on their own that they quickly expanded the lab, laying the foundation to rebuild the company with this new AI. The research group has become a symbol of the company’s greatest ambitions.

Dr. Before being fired, Gebru had sought permission to publish a research paper on how AI-based language systems, including technology developed by Google, could use the prejudiced and hateful language they learn from text in books and websites. Dr. Gebru said he was fed up with Google’s response to such complaints, including its refusal to publish the article.

A few months later, the company announced that Google’s Dr. He fired the other head of the team, Margaret Mitchell, who publicly condemned his handling of the situation with Gebru. Company, Dr. He said you violated Mitchell’s code of conduct.

The paper, published last June in the journal Nature, promoted a technology called reinforcement learning, which it says could improve the design of computer chips. The technology has been hailed as a breakthrough for artificial intelligence and a major improvement over existing approaches to chip design. Google said it used this technique to develop its own chips for artificial intelligence computing.

Google has been working for years to apply its machine learning technique to chip design and has published this. a similar paper one year ago. At the time, Google was working as a research scientist at Intel and a PhD in computer science from the University of California, Berkeley. He asked Chatterjee if the approach could be sold or licensed to a chip design company. said people who knew the subject.

However, Dr. Chatterjee expressed reservations about some of the newspaper’s claims in an internal email, and the three questioned whether the technology had been rigorously tested.

While discussions about this research were ongoing, Google submitted another paper to Nature. For my submission, Google made some edits to the previous article and Dr. He removed the names of two authors who worked closely with Chatterjee and also expressed concern about the article’s main claims.

Some Google researchers were surprised when the new article was published. People say that senior vice president Jeff Dean, who oversees many of the company’s AI efforts, says Dr. He said they believed Gebru did not follow a publication approval process, which he said was necessary following his dismissal.

Google and Anna Goldie, who co-wrote with Azalia Mirhoseini, a computer scientist and one of the paper’s two lead authors, said the changes to the previous paper did not require the full approval process. Google, Dr. He allowed Chatterjee and a handful of internal and external researchers to work on a paper that challenged some of his claims.

The team submitted the rebuttal report to a so-called resolution committee for publication approval. Months later, the paper was rejected.

Researchers working on the rebuttal document said they would like to escalate the matter to Mr. Pichai and Alphabet’s board of directors. They argued that Google’s decision not to publish the rebuttal violated their own decision. artificial intelligence principlesincluding maintaining high standards of scientific excellence. People soon followed Dr. He said he was informed that Chatterjee is no longer an employee.

Miss Goldie, Dr. He said that Chatterjee wanted to manage their project in 2019 and they declined. When she later critiqued she said she failed to substantiate her complaints and ignored the evidence they presented in response.

“Sat Chatterjee has been running a disinformation campaign against me and Azalia for over two years,” Ms Goldie said in a written statement.

He said the study was peer-reviewed by Nature, one of the most prestigious scientific publications. He added that Google is using its own methods to create new chips, and that these chips are currently used in Google’s computer data centers.

Dr. Chatterjee’s lawyer, Laurie M. Burgess, said it was “disappointing that some of the authors of the Nature paper tried to shut down the scientific debate by defaming and attacking Dr. Chatterjee simply for seeking scientific transparency.” Ms. Burgess, one of the 20 co-authors of the journal Nature, Dr. He also questioned Dean’s leadership.

“Jeff Dean’s actions to suppress the publication of all relevant experimental data, not just the data that supports his preferred hypothesis, must be deeply troubling both the scientific community and the wider community that consumes Google services and products,” Ms Burgess said.

Dr. Dean did not respond to a request for comment.

After the rebuttal report was shared with academics and other experts outside of Google, the discussion spread to the global community of researchers specializing in chip design.

Chip maker Nvidia says it uses methods similar to Google’s for chip design, but some experts aren’t sure what Google’s research means for the larger tech industry.

“If this works really well, it would be a really cool thing,” said Jens Lienig, a professor at the Dresden University of Technology in Germany, referring to the AI ​​technology described in Google’s paper. “But it’s unclear whether it works.”