Wednesday, October 21, 2020
Home Lead Story Facebook AI Research Team Develops 'RegNet'

Facebook AI Research Team Develops ‘RegNet’

Facebook AI model beats Google, runs 5 times faster on GPUs

A team from Facebook AI Research (FAIR) has developed a novel low-dimensional design space called ‘RegNet’ that outperforms traditional available models like from Google and runs five times faster on GPUs.

RegNet produces simple, fast and versatile networks and in experiments, it outperformed Google’s SOTA EfficientNet models, said the researchers in a paper titled ‘Designing Network Design Spaces; published on pre-print repository ArXiv. The researchers aimed for “interpretability and to discover general design principles that describe networks that are simple, work well, and generalize across settings”.

Please follow NewsGram on Facebook to get updates on the latest news

The Facebook AI team conducted controlled comparisons with EfficientNet with no training-time enhancements and under the same training setup.

Introduced in 2019, Google’s EfficientNet uses a combination of NAS and model scaling rules and represents the current SOTA.
With comparable training settings and Flops, RegNet models outperformed EfficientNet models while being up to 5× faster on GPUs.

Facebook AI research team recently developed a tool that tricks the facial recognition system to wrongly identify a person in a video. Pixabay

Rather than designing and developing individual networks, the team focused on designing actual network design spaces comprising huge and possibly infinite populations of model architectures. Design space quality is analyzed using error empirical distribution function (EDF).

Analyzing the RegNet design space also provided researchers other unexpected insights into network design. They noticed, for example, that the depth of the best models is stable across compute regimes with an optimal depth of 20 blocks (60 layers).

Please follow NewsGram on Twitter to get updates on the latest news

“While it is common to see modern mobile networks employ inverted bottlenecks, researchers noticed that using inverted bottlenecks degrades performance. The best models do not use either a bottleneck or an inverted bottleneck, said the paper. Facebook AI research team recently developed a tool that tricks the facial recognition system to wrongly identify a person in a video.

The “de-identification” system, which also works in live videos, uses machine learning to change key facial features of a subject in a video. FAIR is advancing the state-of-the-art in artificial intelligence through fundamental and applied research in open collaboration with the community.

Also Read- Can Probiotics Treat Acne? Find it Out Here

The social networking giant created the Facebook AI Research (FAIR) group in 2014 to advance the state of the art of AI through open research for the benefit of all.

Since then, FAIR has grown into an international research organization with labs in Menlo Park, New York, Paris, Montreal, Tel Aviv, Seattle, Pittsburgh, and London. (IANS)



Most Popular

Maa Durga And Cosmic Divinity

By Salil Gewali October is here, and we, with the feeling of enthusiasm, are set to celebrate Durga Puja. For many of us, Durga Puja...

Air Pollution Causes Food Delivery Leading to Plastic Waste

When the air quality outside is bad, office workers are more likely to order food than go out for lunch, which in turn increases...

Virus Can Reinfect COVID Patients When Antibodies Start Depleting

Patients who have recovered from COVID-19 can again get infected by the virus once the antibodies of the viral disease start depleting, the Indian...

International Chef Day a Low-Key Event Due to Pandemic

The International Chef Day is observed on October 20, but this year, the celebrations here were muted as chefs are facing a torrid time...

Coffee Before Breakfast May Increase Diabetes Risk

If you're taking coffee before breakfast, read this carefully. Researchers have found that drinking coffee first thing can have a negative effect on blood...

India’s Nutrition Experts on Supporting Public Health Delivery

Indias public health and nutrition efforts are differentiated by an enterprising early start going as far back as 1975 with the beginning of Anganwadi...

‘Fashion Reboot’- A Collection Focusing Gen Z

The final day of the digital edition of Lotus Make-up India Fashion Week, Spring/Summer 2021 on Sunday opened a film presentation that showcased a...

Disorder in Sleep Cycle Due to Excessive Use of Cellphones

Cell phones have become integral to function for nearly everyone. While it has brought a lot of tasks to our fingertips, cell phones also...

Recent Comments