Machine learning algorithms are generally time consuming specially the training part. However, some applications need the training to be done in real time.
I. DeepLearning Practical Training Course
Time and place: January 15th-January 18th
Two: in-depth intensive study of core technology practical training courses
Time and place: January 27th-January 30th
(Report for three days on the first day; Deploy the computer test in advance)
III. Lecture 1:
Course 1. Introduction to Intensive Learning
Course 2. Strengthen the foundation of learning
Course 3. Deeply strengthen the foundation of learning
Course 4, Multi-agent Deep Intensive Learning
Course 5. Multi-task deep intensive learning
Course 6. Strengthen learning and application
Course 7. Simulation Experiment
Course 8. Supplementary courses
IV. Lecture 2:
Lesson 1: Tensorflow Beginners to Proficiency:
Lesson 2: Image Classification:
Lesson 3: Object Detection:
Lesson 4: Face Recognition:
Lesson 5: Algorithm Implementation:
1. Convolutional neural network CNN
2. Circulating neural network RNN
3. Strengthen learning DRL
4. Antagonistic generation network GAN
5. Transfer learning TL
6. Specific examples of practical medicine, remote sensing images, oil exploration and intelligent transportation
Contact person: Li Luxin (teacher)
Mobile phone: 13311241619
QQ: 1942892948 (same as WeChat)
Strengthen learning QQ exchange group number: 872395038 (add group remarks: Lu xin invited)
Deep learning-distance online course QQ group number: 1057802989 (plus group remarks: Lu xin invited)
There is a variety of machine learning algorithms that are commonly used by the academic community as a benchmark for assessing performance. On the top of the list comes k-Nearest Neighbourhood (k--NN), Support Vector Machine (SVM), and Decision Trees (DT). These algorithms are based on explicit and well-defined known features.
On the other hand, a more sophisticated approach would be to adopt deep learning using Convolutional Neural Network (CNN) which is based on extracted features that are implicit and embedded.
This may vary depending on the size of the data and the number of features the problem contains. There is no single correct answer. It can be found by trying. But the real question is: accuracy or speed, which one is more important?
- Badges
- Science topic