The biggest difference between deep learning and machine learning is "performance"; machine learning is mainly used to make machines possess intelligence, but deep learning is a technology that implements machine learning, and deep learning is also machine learning kind of.
The operating environment of this article: Windows7 system, DELL G3 computer
What is the difference between learning and machine learning?
The biggest difference between deep learning and machine learning is performance.
Machine learning is mainly used to make machines possess intelligence, but deep learning is a technology for realizing machine learning, and deep learning is also a type of machine learning. If the amount of data is relatively small, the performance of deep learning will be relatively poor. This is because deep learning algorithms must have a large amount of data to well understand the patterns.
Generally speaking, artificial intelligence is a relatively topical topic, but now it is still well-known as a field that uses artificial intelligence, and it has had a great impact on these fields. Because of the focus of using artificial intelligence, systems have been developed that can not only simulate human thought processes, but also learn knowledge from processing data, and this phenomenon is machine learning.
#1. Data dependence. The main difference between deep learning and machine learning is performance. When the amount of data is small, the performance of deep learning is not good, because deep learning algorithms require a large amount of data to well understand the patterns contained in it.
2. Hardware support. Deep learning algorithms rely heavily on high-end machines, while traditional machine learning algorithms can run on low-end machines. Deep learning requires GPUs to do a lot of matrix multiplication operations.
3. Feature engineering. Feature engineering is to input domain knowledge into the feature extractor to reduce data complexity. This process is very expensive in terms of time and expertise.
4. Solution, usually, we use traditional algorithms to solve problems. This requires breaking the problem into parts, solving them separately, and then combining them after getting the results.
5. Execution time, because deep learning contains a lot of parameters, it will take more time than machine learning. Machine learning takes less time to train data, taking only seconds to hours.
The main application scenarios are:
Computer vision: license plate recognition, face recognition.
Information retrieval: search engine, text retrieval, image retrieval.
Marketing: automatic email marketing, target identification.
Medical Diagnosis: Cancer Detection, Anomaly Detection.
Natural language processing: semantic analysis, photo tagging, online advertising.
If we look at the outlook, the main points are:
1. Machine learning and data science are gaining momentum, and using machine learning in their business is becoming increasingly important for companies that want to survive.
2. Deep learning has proven to be one of the most advanced technologies in existence. It has brought countless surprises to people, and I believe it will do so in the future.
3． Researchers are still exploring machine learning and deep learning. In the past, research on the two was limited to academic scope, but now the industry has also increased its research efforts.
The best proof is image recognition, which is increasingly becoming an area led by AI. The system can be designed to manipulate pre-written routines that analyze shapes, colors and objects in pictures, scanning millions of images in order to teach itself how to correctly identify images.
For more related knowledge, please visit the FAQ column!
The above is the detailed content of What is the difference between deep learning and machine learning. For more information, please follow other related articles on the PHP Chinese website!