Deep Learning

Deep learning is a branch of machine learning that trains neural networks with many stacked layers. Each layer learns progressively more abstract features, from edges in an image to whole objects, or from letters to meaning.

Why it matters

Deep learning drives modern speech recognition, image classification, translation, and large language models. It removed much of the manual feature engineering older methods required, though it typically needs large datasets and substantial compute, usually GPUs.

Example

An image classifier might pass a photo through dozens of layers. Early layers detect edges, middle layers detect shapes like wheels, and the final layer outputs probabilities.

input: photo.jpg
output: car 0.94, truck 0.04, bus 0.01

Common misconceptions

• Deep learning is always better. On small tabular datasets, gradient-boosted trees or linear models often win and are cheaper.
• Deeper means smarter. Extra layers raise cost and can worsen overfitting without enough data.
• It is a black box that cannot be examined. Interpretation tools exist, but explanations remain imperfect.

Related

• Neural network
• Machine learning
• GPU
• Transformer

यह पेज सामान्य जानकारी है, जिसे हमारी संपादकीय टीम ने लिखा है और दस्तावेज़ीकरण बदलने पर अपडेट किया जाता है। सर्वर और प्रदाता अलग-अलग होते हैं, इसलिए बदलाव पहले सुरक्षित तरीके से जाँचें। देखें: संपादकीय नीति और अस्वीकरण.