Want smarter insights in your inbox? Sign up for our weekly newsletters to get only what matters to enterprise AI, data, and security leaders. Subscribe Now
Facebook today announced it plans to open-source some of its AI tools, including Translate, which translates 48 languages, and ELF, which teaches machines reasoning through gameplay. Facebook’s AI now conducts more than 6 billion translations a day. An ELF bot has competed with some of the top Go players around the world in recent weeks and currently has a record of 14-0, said Facebook CTO Mike Schroepfer.
“Developers can also take advantage of tools like Glow, a machine learning compiler that accelerates framework performance on different hardware platforms, and Tensor Comprehensions, a tool that automatically generates efficient GPU code from high-level mathematical operations,” the company said in a blog post. “We have also open-sourced other libraries, such as Detectron, which supports object-detection research, covering both bounding box and object instance segmentation outputs.”
Above: Facebook’s Go playing bot on display
Plans to release PyTorch 1.0 in the coming months were also announced today.
The open source deep learning framework, first released to the public in January 2017, has been downloaded more than 1.1 million times and was the second-most cited deep learning framework on popular research portal Arxiv over the past month.
AI Scaling Hits Its Limits
Power caps, rising token costs, and inference delays are reshaping enterprise AI. Join our exclusive salon to discover how top teams are:
- Turning energy into a strategic advantage
- Architecting efficient inference for real throughput gains
- Unlocking competitive ROI with sustainable AI systems
Secure your spot to stay ahead: https://bit.ly/4mwGngO
ONNX, an open source project begun by Facebook, Amazon, and Microsoft last fall to make it easier to share information among open source systems like Caffe2 and PyTorch Microsoft’s Cognitive Toolkit, will play a larger role in the new edition of PyTorch. ONNX will act as the model export format in PyTorch 1.0 and will allow for the integration of accelerated runtimes or hardware-specific libraries.
Microsoft’s Azure machine learning services and Amazon Web Services will both support PyTorch 1.0.
The news was shared onstage at F8, Facebook’s annual developer conference taking place yesterday and today, May 1 and 2, at McEnery Convention Center in San Jose, California. Other news announced at F8 this year include the release of Oculus Go, new Facebook Stories sharing capabilities, and the reopening of app and bot reviews following the Cambridge Analytica scandal. See the full rundown here.
