Deep learning is fairly new and is defined as a subset of machine learning that achieves great power and flexibility by learning to represent the world as nested hierarchy of concepts, with each concept defined in relation to simpler concepts, and more abstract representations computed in terms of less abstract ones. Clear as mud? This interesting topic came to us from IDG Connect in their article, “The growth of deep learning.”

Deep learning is quickly overtaking more classic machine learning methods. A big advantage with deep learning, and a key part in understanding why it’s becoming popular, is that it’s powered by massive amounts of data.

Deep learning requires high-end machines contrary to traditional machine learning algorithms. It has made several breakthroughs over the past decade and identifying patterns and classifying image data. This has been particularly helpful to the healthcare sector, where it can be used to analyze scans.

Melody K. Smith

Sponsored by Data Harmony, a unit of Access Innovations, the world leader in indexing and making content findable.