Deep learning
A type of computer program that learns to recognize patterns by using many layers of artificial 'neurons' to process information, similar to how a human brain learns to identify complex objects by looking at smaller details first.
A subset of machine learning that utilizes deep neural networks with multiple hidden layers to automatically extract and learn hierarchical representations from raw data, effectively automating the feature engineering process.
Machine learning using neural networks with many layers, which learn hierarchical feature representations directly from raw data rather than relying on hand-engineered features.
evolution
- 1986 · historyBackpropagation Popularization
Rumelhart, Hinton, and Williams popularized the backpropagation algorithm, enabling the training of multi-layer neural networks.
- 1998 · historyLeNet-5
Yann LeCun introduced the LeNet-5 convolutional neural network, demonstrating effective multi-layer learning for handwritten digit recognition.
- 2012 · historyAlexNet Breakthrough
Alex Krizhevsky and colleagues won the ImageNet competition by a large margin using a deep convolutional neural network trained on GPUs.
- 2017 · historyTransformer Architecture
Google researchers introduced the Transformer model, shifting deep learning toward attention-based mechanisms that power modern large language models.