In CRYPTO 2019, Gohr first introduced the deep learning method to cryptanalysis for SPECK32/64. A differential-neural distinguisher was obtained using ResNet neural network. Zhang et al. used multiple parallel convolutional layers with different kernel sizes to capture information from multiple dimensions, thus improving the accuracy or obtaining a more round of distinguisher for SPECK32/64 and SIMON32/64. Inspired by Zhang's work, we apply the network structure to other ciphers. We not only improve the accuracy of the distinguisher, but also increase the number of rounds of the distinguisher, that is, distinguish more rounds of ciphertext and random number for DES, Chaskey and PRESENT.

In recent years, algebraic attacks and fast algebraic attacks have received a lot of attention in the cryptographic community. There are three Boolean functions achieving optimal algebraic immunity based on primitive element of F2n. The support of Boolean functions in [1]-[3] have the same parameter s, which makes us have a large number of Boolean functions with good properties. However, we prove that the Boolean functions are affine equivalence when s takes different values.