The following operation rules are wrong: ()
A. In implication operation, if A is true, when B is true, A → B is true, B is false, a → B is false.
B. In implication operation, if A is false, then whether B is true or false, a → B is true.
C. In the absorption operation, A ∨ (A ∧ B) = A, A ∧ (A ∨ B) = B.
D. In the complementary law, A∧ (-A) = 0, a∨ (-A) = 1.
In the following statement about brain computer interface, the wrong one is ( )
A. Brain computer interface system generally includes two parts: signal acquisition and signal processing
Brain computer interface (BCI) is a kind of communication control system established by EEG signal between brain and external equipment
C. Brain computer interface system depends on the communication between nervous system and muscle tissue
D. Signal acquisition is mainly used to filter EEG signals and eliminate interference noise
() is false statement about CNN.
Artificial neural network is mainly composed of input layer, hidden layer and output layer.Deep learning generally requires multiple hidden layers.
B. The convolutional neural network has the characteristics of natural network weight sharing and local sparse network connection, which also coincides with the sparse response characteristic of biological neurons.
Convolutional neural network has two basic concepts: weight sharing and pooling.Weight sharing reduces the number of weight parameters.Pooling can reduce the characteristic graph and simplify the calculation.
D. The function of loss function layer is used to estimate the difference between the predicted value and the actual value of the model. It is a non-negative real numerical function.
() is false statement about DNN.
A. RNN: Speech recognition, question and answer system, language modeling and text generation, etc.
B. Gan: Image synthesis, semantic image editing, style migration, image super-resolution technology and classification.
C. VAE: From generating fake faces to composing music, etc.
D. LSTM: The long distance historical information in the sequence is captured, but the long-term dependency problem cannot be alleviated. RNN should be used to effectively alleviate the long-term dependency.
() is true statement about history of Artificial intelligence.
A. In 1949, psychologist Herbert Hebb proposed the Hebb rule in his book Behavior Self-Organization.
B. In 1957, Rosenblat defined a neural network structure called the perceptron.
C. In 1982, physicist Hopfield proposed a fully connected, discrete neural network model.This is a new neural network model with complete theoretical basis.
D. In 1986, a parallel computing research group in the United States proposed the back-propagation (BP) learning algorithm of the aforementioned feedback neural network.
微信支付 
支付宝支付