comscore Facebook AI model tops Google, runs 5 times faster on GPUs | BGR India
News

Facebook's RegNet AI model tops Google's EfficientNet, runs 5 times faster on GPUs

The new RegNet model by Facebook is developed by a team from Facebook AI Research (FAIR).

  • Published: April 12, 2020 3:40 PM IST
facebook-pixabay-main

Source: Pixabay


Facebook recently made progress in the GPU department. A team from Facebook AI Research (FAIR) recently developed a new low-dimensional design space. Named ‘RegNet’ the new design outperforms traditional available models including ones from Google. Further, it runs five times faster on GPUs. Also Read - Facebook launches Quiet Mode to curb social media addiction; here's when you will get it

RegNet produces simple, fast and versatile networks. Moreover, in certain experiments, it even outperformed Google’s SOTA EfficientNet models, said researchers in a paper titled ‘Designing Network Design Spaces. The same can also be found 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”. Also Read - Facebook quietly adds private dating app to App store

Watch: 5 Tips to Save Mobile Data

The Facebook AI team also conducted controlled comparisons with Google’s EfficientNet with no training-time enhancements, under the same training setup. Introduced in 2019, Google’s EfficientNet design uses a combination of NAS and model scaling rules, representing the current SOTA. With similar training settings and Flops, RegNet models outperformed EfficientNet models also being up to 5 times faster on GPUs. Also Read - Facebook tried to buy NSO Spyware Pegasus to monitor its users: Report

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

Further analyzing RegNet’s design space also gave researchers other unexpected insights into its network design. For instance, they noticed that the depth of the best models is stable across regimes with an optimal depth of 20 blocks (60 layers).

“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’s AI research team recently developed a tool that tracks the facial recognition system to wrongly identify people in video footage. The “de-identification” system, which also works in live videos, uses machine learning to change key facial features of a subject in real-time. FAIR is advancing the state-of-the-art in artificial intelligence through fundamental and applied research in open collaboration with the community.

The history behind FAIR

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.

(With inputs from IANS)

For the latest tech news across the world, latest PC and Mobile games, tips & tricks, top-notch gadget reviews of most exciting releases follow BGR India’s Facebook, Twitter, subscribe our YouTube Channel.
  • Published Date: April 12, 2020 3:40 PM IST



new arrivals in india

Poco X3
Poco X3

16,999

Realme Narzo 20A
Realme Narzo 20A

8,499

Realme Narzo 20
Realme Narzo 20

10,499

Realme Narzo 20 Pro
Realme Narzo 20 Pro

14,999

Oppo F17
Oppo F17

17,990

Samsung Galaxy M51
Samsung Galaxy M51

24,999

Poco M2
Poco M2

10,999

Oppo F17 Pro
Oppo F17 Pro

22,990

Realme 7 Pro
Realme 7 Pro

19,999

Realme 7
Realme 7

14,999

Xiaomi Redmi 9A
Xiaomi Redmi 9A

6,799

Vivo Y20
Vivo Y20

12,990

Xiaomi Redmi 9
Xiaomi Redmi 9

8,999

Nokia 5.3
Nokia 5.3

13,999

Motorola Moto G9
Motorola Moto G9

11,499

Realme C15
Realme C15

9,999

Realme C12
Realme C12

8,999

Samsung Galaxy Note 20
Samsung Galaxy Note 20

77,999

Xiaomi Redmi 9 Prime
Xiaomi Redmi 9 Prime

9,999

Oppo Reno4 Pro
Oppo Reno4 Pro

34,990

Samsung Galaxy M01 Core
Samsung Galaxy M01 Core

5,499

Realme 6i
Realme 6i

12,999

Asus Rog Phone 3
Asus Rog Phone 3

49,999

OnePlus Nord
OnePlus Nord

24,999

Infinix Smart 4 Plus
Infinix Smart 4 Plus

7,999

Xiaomi Redmi Note 9
Xiaomi Redmi Note 9

11,999

Samsung Galaxy M01s
Samsung Galaxy M01s

9,999

Vivo X50 Pro 5G
Vivo X50 Pro 5G

49,990

Vivo X50 5G
Vivo X50 5G

34,990

Realme C11
Realme C11

7,499

Poco M2 Pro
Poco M2 Pro

13,999

Realme X3
Realme X3

24,999

Realme X3 SuperZoom
Realme X3 SuperZoom

27,999

Tecno Spark Power 2
Tecno Spark Power 2

9,999

Oppo A12
Oppo A12

9,990

Oppo A52
Oppo A52

16,990

Samsung Galaxy A21s
Samsung Galaxy A21s

15,999

Oppo Find X2
Oppo Find X2

64,990

Motorola One Fusion Plus
Motorola One Fusion Plus

17,499

Samsung Galaxy A31
Samsung Galaxy A31

20,999

Samsung Galaxy M01
Samsung Galaxy M01

8,999

Samsung Galaxy M11
Samsung Galaxy M11

10,999

Infinix Hot 9 Pro
Infinix Hot 9 Pro

9,999

LG Velvet
LG Velvet

Price Not Available

Xiaomi Mi Note 10 Lite
Xiaomi Mi Note 10 Lite

Price Not Available

Apple iPhone SE 2020
Apple iPhone SE 2020

42,500

Honor 30 Pro
Honor 30 Pro

Price Not Available

Honor 30
Honor 30

Price Not Available

OnePlus 8
OnePlus 8

44,999

OnePlus 8 Pro
OnePlus 8 Pro

54,999

Xiaomi Redmi Note 9 Pro
Xiaomi Redmi Note 9 Pro

13,999

Motorola Moto E4
Motorola Moto E4

8,999

Samsung Galaxy On Max
Samsung Galaxy On Max

9,775

nubia N2
nubia N2

15,999

Karbonn K9 Kavach 4G
Karbonn K9 Kavach 4G

5,290

Motorola Moto C Plus
Motorola Moto C Plus

6,999

Best Sellers