MENU × BUSINESS
Banking And InsuranceCryptocurrencyDigital MarketingErpFood And BeveragesHealthcareLegalMarketing And AdvertisingMedia And EntertainmentMetals And MiningOil And GasRetailTelecom
TECHNOLOGY
Artificial IntelligenceBig DataCloudCyber SecurityE CommerceEducationGaming And VfxIT ServiceMobileNetworkingSAPScience And TechnologySecuritySoftwareStorage
PLATFORM
CiscoDatabaseGoogleIBMJuniperM2MMicrosoftOracleOracleRed Hat
LEADERSHIP
CEO ReviewCompany Review
MAGAZINE
ASIA INDIA
STARTUPS CLIENT SPEAK CONTACT US

The Silicon Review Asia

Microsoft and Facebook have announced Open Neural Network Exchange

Microsoft and Facebook have announced Open Neural Network Exchange

Microsoft and Facebook have announced a new open source project today that’s aimed at creating a shared representation for neural networks across different programming frameworks. Called the Open Neural Network Exchange (ONNX), the new project will make it possible to share models across the Cognitive Toolkit, PyTorch, and model Caffe2.

Microsoft, which last week announced that it was partnering with Amazon to let their respective AI-powered virtual assistants talk, said this week it has joined forced with Facebook to launch an open source AI resource repository called Open Neural Network Exchange (ONNX).

The system works by tracing how a neural network generated using one of these frameworks executes at runtime and then using that information to create a generic computation graph that can be moved around. That’s possible because each of those frameworks produces a very similar end result when it comes to computation, even though the higher level representation is different.

Right now, the biggest issue with ONNX is that it isn’t compatible with some other popular machine learning frameworks, including TensorFlow, which originated at Google, and Apache MXNet, which is Amazon’s preferred machine learning framework.

"People experimenting with new models, and particularly those in research, want maximum flexibility and expressiveness in writing neural networks - ranging from dynamic neural networks to supporting gradients of gradients, while keeping a bread-and-butter ConvNet performance", developers at Facebook wrote.

"Researchers also want to iterate rapidly, which means that they need excellent tooling for interactive development and debugging. PyTorch has been built to push the limits of research frameworks, to unlock researchers from the constraints of a platform and allow them to express their ideas easier than before."

YOU MAY ALSO LIKE

Salesforce Expands into Thailand, Opens Office in Bangkok

Salesforce has announced permanent presence in Thailand by launching its new office in the country’s capital, Bangkok. The permanent presence wi...

GMLL is expected to soon add two more stores to its retail arm ‘Price Mantra’

Garment Mantra Lifestyle, a popular name in the Indian fashion retail segment, recently made an announcement that the company is expanding its retail ...

Tariff plans will cost higher from the next financial year as telecom companies are gearing up to increase rates

The ongoing Covid pandemic had significantly increased the number of mobile and internet users worldwide. The high amount of usage is expected to drop...

Business Travelers to Stay at Singapore’s Changi Airport Bubble

Singapore’s open economy depends heavily on tourism and business. But its vibrant but small economy has been hurting as the circumstances due to...

RECOMMENDED