Deep thinking

摘要

Artificial intelligence (AI) is an umbrella term for the methodologies used to make machines behave in a way we would consider'smart. Deep learning is a method of AI, and a subset of machine learning, that enables computers to imitate human learning. The recycling industry continues to be transformed by deep learning because it enables humans' cognitive abilities to be replicated on machines or computers for use in sorting tasks. In classical machine learning, an expert supplies the algorithm with only the most useful data it requires to enable it to separate different materials. In deep learning, the algorithm can be trained using the raw data - thousands of images, for example - to identify material that needs to be separated during the sorting process. Tomra Sorting has used AI in its machines since the early days of sorting, but the technology has evolved significantly over the years. Today, by mimicking the activity of large numbers of layers of neurons in the human brain, deep-learning algorithms can perform clearly defined, complex cognitive tasks in the same way as humans - but much faster and, in some cases, with much greater accuracy.