In the research of deep learning, an Unsupervised deep convolution-generated Generated Adversarial Network (UGAN) usually needs a large number of data samples to train. However, when faced with some small samples, the performance of the algorithm is often degraded due to over-fitting. Combined with specially designed data enhancement methods, a generated adversarial network optimization algorithm based on adaptive data augmentation (AdauGAN) is proposed. The adaptive data augmentation module is added before the discriminant network, and a spatial transformation is carried out simultaneously at the probability distribution level of generated data and real data. To alleviate the over-fitting phenomenon in the training process, the current enhancement intensity is adjusted adaptively after the over-fitting occurs. The proposed algorithm is verified on SVHN, CelebA and CIFAR-10 data sets. The Frechet Inception Distance (FID) values of AdauGAN achieve 22.10, 23.94, 34.87, respectively, which is close to or even higher than the training results of Deep Convolution Generated Adversarial Network (DCGAN) under all data. Extensive experiment results show that the proposed Adaugan has an excellent performance in small samples. Besides, in some cases, it can catch up with the large sample results of existing algorithms.

Federated learning is a distributed machine learning method that allows numerous clients to cooperatively train AI models without uploading their raw data. However, the heterogeneous data will significantly affect the performance of the global model, and self-interested clients are often reluctant to participate in learning tasks without satisfactory rewards. To tackle these challenges, this paper proposes a novel incentive mechanism, called Qualiaty-Aware Reverse Auction (QARA), to ensure that clients can maximize their own utilities when conducting honest bidding. Specifically, we first use Shannon entropy to combine data quantity and diversity to measure the quality of client data, allowing clients to be selected and incentivized based on their data quality and bids within a limited budget. Next, we formalize the data quality maximization problem with the aim of maximizing the quality of the selected client data, and show that it is NP-hard. Furthermore, to solve such a intractable problem, we design a greedy algorithm with an approximation ratio of $1 / 2$. Theoretically, we prove that QARA satisfies dominant strategy incentive compatibility, computational efficiency, individual rationality, and budget feasibility. Last, we evaluate the generalization of QARA and four baselines on five real-world datasets to demonstrate the superiority of QARA.

Few-shot learning can solve new learning tasks in the condition of fewer samples. However, currently, the few-shot learning algorithms mostly use the ResNet as a backbone, which leads to a large number of model parameters. To deal with the problem, a lightweight backbone named DenseAttentionNet which is based on the Convolutional Block Attention Mechanism is proposed by comparing the parameter amount and the accuracy of few-shot classification with ResNet-12. Then, based on the DenseAttentionNet, a few-shot learning algorithm called Meta-DenseAttention is presented to balance the model parameters and the classification effect. The dense connection and attention mechanism are combined to meet the requirements of fewer parameters and to achieve a good classification effect for the first time. The experimental results show that the DenseAttentionNet, not only reduces the number of parameters by 55% but also outperforms other classic backbones in the classification effect compared with the ResNet-12 benchmark. In addition, Meta-DenseAttention has an accuracy of 56.57% (5way-1shot) and 72.73% (5way-5shot) on the miniImageNet, although the number of parameters is only 3.6[Formula: see text]M. The experimental results also show that the few-shot learning algorithm proposed in this paper not only guarantees classification accuracy but also has the characteristics of lightweight.