Lesson 7 How do machines learn our preferences?
How does a machine that can't communicate understand your interests and recommend products and videos for you?
Lesson 8 The Turing Test: Can Machines Think?
Can machines think? What are "template matching" and "weird talk"? Why are they models for dialogue systems?
Lesson 9 Find sweet strawberries in a few steps
How many steps does it take to train a robot that can tell whether strawberries are sweet or not?
First use 6 strawberries to train several robots that can judge whether strawberries are sweet or not. After training, use two strawberries to test their situation, leaving the robot with the most accurate judgment; finally take out two strawberries that he has never seen before. Check his performance.
Before training the model, we divide the collected raw data into three parts: training set, validation set, and test set. The training set is used to build the model; after obtaining multiple models, the validation set is used to find the optimal model among them; finally, the test set is used to test the performance of the optimal model to ensure that it is truly usable.
Lesson 10 What is a Knowledge Graph
In addition to turning into a cold string, how to better map the real world to the data world? What is a knowledge graph? What do nodes represent? What does side represent?
Lesson 11 Knowledge Graph of Artificial Intelligence Proper Terms
What is the relationship between artificial intelligence, machine learning, and deep learning? After finally understanding the meaning of a large number of artificial intelligence terms, but do not know their position in the entire artificial intelligence system?
Lesson 12 What is Language?
Why is natural language processing the "jewel in the crown of artificial intelligence"? What does this have to do with language itself?
Language is made up of sentences, sentences are made up of phrases, phrases are made up of words, and words are made up of morphemes. A morpheme is the smallest meaningful unit of language. The smallest unit of spoken language is a phoneme. Choosing a phoneme as the modeling unit for language recognition can avoid data sparsity caused by words as units. Morphemes can help natural language comprehension to better understand the meaning of words.