Which of the following statements about brain-like computing development level is not correct? ()
A. Brain-like computers and brain-like robots belong to the product layer.
B. Machine vision, auditory perception, multimodal collaborative perception, natural language understanding, etc. belong to the software layer.
C. Pulse neural networks, brain plasticity mechanisms, brain maps, etc. belong to the basic theoretical layer.
D. Brain-like chips belong to the hardware layer.
查看答案
Which following is wrong about the characteristics of reinforcement learning
A. Quick and prompt feedback
B. No supervision, only reward
C. Strong versatility and popularization, the agent is also applied in a new environment of true uncertainty.
D. The behavior of the agent will affect the subsequent data and have a continuous impact on the environment
What is correct for deep learning is
A. By building a multi-hidden layer model and massive training data to learn more useful features, and ultimately improve the accuracy of classification or prediction
B. "Deep model" is the means, "feature learning" is the purpose
C. Emphasizes the depth of the model structure, usually there are 5-10 layers of hidden nodes
D. Compared with the method of artificially constructing features, using big data to learn features is more able to characterize the rich internal information of the data
Natural language understanding is an important application field of artificial intelligence () in the following list is not its goal.
A. Understand what others say.
B. Analyze, summarize or edit the information expressed in natural language.
C. enjoy music.
D. Machine translation.
The data required during the processing of an intelligent machine includes:（）
A. Performance evaluation data
B. Sensor data during processing
C. Inspection data during processing
D. Inspection data after processing
微信支付 
支付宝支付