DynGAN: Solving Mode Collapse in GANs With Dynamic Clustering.

Generative Adversarial Networks (GANs) are widely-used generative models for synthesizing complex and realistic data. However, mode collapse, where the diversity of generated samples is significantly lower than that of real samples, poses a major challenge for further applications. Our theoretical analysis demonstrates that the generator loss function is non-convex with respect to its parameters when there are multiple modes in real data. In particular, parameters that result in generated distributions with perfect partial mode coverage of the real distribution are the local minima of the generator loss function. To address mode collapse, we propose a unified framework called Dynamic GAN. This method detects collapsed samples in the generator by thresholding on observable discriminator outputs, divides the training set based on these collapsed samples, and trains a dynamic conditional model on the partitions. The theoretical outcome ensures progressive mode coverage and experiments on synthetic and real-world data sets demonstrate that our method surpasses several GAN variants. In conclusion, we examine the root cause of mode collapse and offer a novel approach to quantitatively detect and resolve it in GANs.

Multi-resolution auto-encoder for anomaly detection of retinal imaging.

Identifying unknown types of diseases is a crucial step in preceding retinal imaging classification for the sake of safety, which is known as anomaly detection of retinal imaging. However, the widely-used supervised learning algorithms are not suitable for this problem, since the data of the unknown category is unobtainable. Moreover, for retinal imaging with different types of anomalous regions, using a single-resolution input causes information loss. Therefore, we propose an unsupervised auto-encoder model with multi-resolution inputs and outputs. We provide a theoretical understanding of the effectiveness of reconstruction error and the improvement of self-supervised learning for anomaly detection. Our experiments on two widely-used retinal imaging datasets show that the proposed methods are superior to other methods, and further experiments verify the validity of each part of the proposed method.

Active learning with label quality control.

Training deep neural networks requires a large number of labeled samples, which are typically provided by crowdsourced workers or professionals at a high cost. To obtain qualified labels, samples need to be relabeled for inspection to control the quality of the labels, which further increases the cost. Active learning methods aim to select the most valuable samples for labeling to reduce labeling costs. We designed a practical active learning method that adaptively allocates labeling resources to the most valuable unlabeled samples and the most likely mislabeled labeled samples, thus significantly reducing the overall labeling cost. We prove that the probability of our proposed method labeling more than one sample from any redundant sample set in the same batch is less than 1/k, where k is the number of the k-fold experiment used in the method, thus significantly reducing the labeling resources wasted on redundant samples. Our proposed method achieves the best level of results on benchmark datasets, and it performs well in an industrial application of automatic optical inspection.

Biology-Informed Recurrent Neural Network for Pandemic Prediction Using Multimodal Data.

In the biomedical field, the time interval from infection to medical diagnosis is a random variable that obeys the log-normal distribution in general. Inspired by this biological law, we propose a novel back-projection infected-susceptible-infected-based long short-term memory (BPISI-LSTM) neural network for pandemic prediction. The multimodal data, including disease-related data and migration information, are used to model the impact of social contact on disease transmission. The proposed model not only predicts the number of confirmed cases, but also estimates the number of infected cases. We evaluate the proposed model on the COVID-19 datasets from India, Austria, and Indonesia. In terms of predicting the number of confirmed cases, our model outperforms the latest epidemiological modeling methods, such as vSIR, and intelligent algorithms, such as LSTM, for both short-term and long-term predictions, which shows the superiority of bio-inspired intelligent algorithms. In general, the use of mobility information improves the prediction accuracy of the model. Moreover, the number of infected cases in these three countries is also estimated, which is an unobservable but crucial indicator for the control of the pandemic.