Enot.ai Introduces Solution to Optimize Deep Neural Networks

Missed a session at the Data Summit? View on demand here.

Latvian headquarters Enot, an AI startup backed by New Nordic Ventures, has announced the launch of a special framework to optimize deep neural networks.

The AI-driven offering, as the company explains, comes with a Python API that can be integrated into various neural network training pipelines. It then automates the search for an optimal network architecture, taking into account various hardware and software-oriented parameters, including RAM, latency, model size constraints, and type of operation.

“Enot.ai’s framework takes a trained neural network as input, then selects a subnetwork with the lowest latency that can guarantee no degradation in accuracy,” the company told VentureBeat.

The whole process makes the neural network faster, smaller and more energy efficient, solving the challenges developers face worldwide.

The impact of Enot

Enot claims its solution has the potential to help AI developers and businesses achieve up to 20 times neural network acceleration and up to 25 times network compression. The benefits even help reduce computer hardware costs by as much as 70%.

“Enot is at the forefront of next-level AI optimization, helping to bring rapid, real-time levels of AI progress… Our journey has only just begun with examples such as the Weedbot laser weeding machine becoming 2.7 times faster, thanks to the Enot framework,” said Sergey Aliamkin, CEO and founder of the company.

Overall, the company claims to have piloted projects with more than 20 companies, including major players such as PicsArt, LG, Huawei, Dscribe, and Hive.aero.

In one case, it accelerated an image enhancement neural network by 13.3 times for a smartphone manufacturer. The optimization reduced the depth of the neural network from 16 to 11 and reduced the input resolution from 224 x 224 pixels to 96 x 96 pixels, without any loss of accuracy, the company said. It also had another project with the same company, where the framework delivered 5.1-fold acceleration for a photo-noise reduction neural network, without any quality change.

“Before meeting us, they already had several clients, including major international technology companies such as LG, Huawei, Sony. That confirmed to us that Enot is solving a mission-critical problem in the neural network space that cannot be solved internally, nor are there any viable solutions on the market,” Dmitry Saikovsky, General Partner of New Nordic Ventures, told Venturebeat.

Other players looking to solve the same problem include Deci.ai, OctoML, DeepCube, Deeplite, NeuralMagic, and DarwinAI.

VentureBeat’s mission is to be a digital city square for tech decision makers to learn about transformative business technology and transactions. Learn more

This post Enot.ai Introduces Solution to Optimize Deep Neural Networks

was original published at “https://venturebeat.com/2022/03/11/enot-ai-debuts-solution-to-optimize-deep-neural-networks/”