A new study shows that deep learning models can learn to recognize handwritten digits from scratch

Deep learning has been one of the most successful areas of artificial intelligence research in recent years. These algorithms can automatically extract features from complex data and use them to make predictions with incredible accuracy. But what happens if you want to train a deep learning model to recognize something entirely new? A new study published in the journal Nature Communications shows that it is possible for deep learning models to learn to recognize handwritten digits from scratch.