The Future of Deep Learning According to Its Pioneers
In the past five years, the number of people working on deep learning has exploded. In that time, the number of researchers has increased from a handful to tens of thousands. Moreover, the pace of research has accelerated and there are no signs of this trend slowing down any time soon.
3 min readAug 27, 2022
Self-supervised learning
Self-supervised learning is a technique in machine learning that trains a model to recognize part of an input from another part. It is also known as predictive learning or pretext learning. The technique turns an unsupervised problem into a supervised one by auto-generating labels. The key to this technique is setting the right learning objectives. The goal is to train a system for a specific use case.
As the field of artificial intelligence continues to grow, many advances have been made. For example, self-supervised learning is being used in computer vision, cancer diagnosis, and some areas of natural language processing. This technique has also found a place in sensitive applications such as evaluating content on social media.
