"The largest models take a million dollars to train, they are the equivalent of a supercomputer," says Intel's GM for AI and analytics, Wei Li. "That's not sustainable, it's not going to be widely applicable if everybody has to pay ten million dollars to train a trillion-parameter model." A consensus has emerged in deep learning circles of artificial intelligence about a couple of basic notions. One notion is that the trend for neural networks to get bigger and bigger will continue for the foreseeable future, with some exceptions. Another consensus view is that the potential of deep learning rests on software innovation, despite a proliferation of novel hardware. Taken together, what do those two things mean?

The challenges associated with getting AI models into production are well known. What developers need now are ways to address those challenges. Ramtin Davanlou, from Accenture*, presents his thoughts around how developers can get their models in production and do it more quickly during his presentation at the oneAPI DevSummit for AI 2022. Davanlou first discusses the development cycle to deploy AI systems and its associated challenges. He emphasizes that to move through this cycle successfully requires the combination of both hardware acceleration and the software optimizations provided by oneAPI.

There is an ever-growing number of accelerators in the world. This raises the question of how various ecosystems will evolve to allow programmers to leverage these accelerators. At higher levels of abstraction, domain-specific layers like Tensorflow and PyTorch provide great abstractions to the underlying hardware. However, for developers who maintain code that talks to the hardware without such an abstraction the challenge still exists. One solution that is supported on multiple underlying hardware architectures is C with SYCL.

Intel Experience Day 2019, organized by Intel, one of the major innovative hardware and technology corporations worldwide, took place in Moscow at the end of October. Intel and partner companies presented the latest Intel hardware and software product implementations advancing IoT, AI, computer vision, machine learning, object recognition, and more. Many speakers shared their ideas and insights on trending industrial innovations like cloud computing, Big Data, and analytics, including Al Diaz, Intel's Vice President, Natalya Galyan, Intel's Regional Director for Russia, and Marina Alekseeva, CEO of R&D of Intel in Russia. Intel Experience Day 2019 attracted many IT market players who use Intel solutions in their work daily, and Auriga experts were among them. Several years ago, Auriga became a pioneer user of the Intel Multi-OS Engine tool to develop an innovative iPad application for patient monitoring.