There were a number of presentations on artificial intelligence (AI) at the year's Intel Innovation. It includes this video about Increasing productivity and performance using the Intel Neural Compressor (INC), an open-source Python library. INC helps developers quickly deploy low-precision inference solutions on popular deep-learning frameworks. Developers using INC can "increase their productivity by up to 18X and increase performance by up to 30X when quantizing models from FP32 to INT8 numerical formats."
