Search

Variational autoencoder is a commonly unsupervised learning model. However, its complex structure hinders the utilization of the network architecture search algorithm to release researchers from tedious manual design. To design excellent architectures automatically, this paper proposes an efficient evolutionary architecture search for variational autoencoder with alternating optimization and adaptive crossover(AOC-VAE). Firstly, to alleviate the problem of large search space when automatically designing variational autoencoders, AOC-VAE designs an alternating optimized search mechanism based on the specific coupling of encoder and decoder in variational autoencoders, which reduces the original huge search space almost to half. Then, AOC-VAE can find quickly the optimal individual in the solution space by designing an adaptive crossover mechanism. In early evolutionary period, the structural differences between individuals are relatively significant, making crossover operations more inclined to exchange structural information between individuals. As evolution progresses, the individual structures in the population tend to be similar, and the exchange of information concentrates on the parameter. Finally, in the optimization process, a fitness evaluation mechanism based on dynamic weights is designed to accurately find out the outstanding individuals under the current optimization goal. Individual fitness in the population is more inclined to be affected by the current optimization goal, thus guiding the population to evolve according to the optimization goal at different stages. AOC-VAE is verified on MNIST, SVHN, CIFAR-10, and CIFAR-100 benchmark datasets and compared with 14 algorithms. The experimental results show that the VAE network structure designed by the AOC-VAE performs well in the image classification task. [ABSTRACT FROM AUTHOR]

Evolutionary neural architecture search (ENAS) algorithms benefit from the non-convex optimization capability of evolutionary computation. However, most ENAS algorithms use a genetic algorithm (GA) with fixed parameter settings, which limits the algorithm's search performance. Furthermore, the information generated during the evolutionary process is often ignored, while this information can be helpful for guiding the evolutionary direction of the population. This paper proposes a novel ENAS algorithm based on adaptive parameter control and gene potential contribution (AG-ENAS) to evolve neural networks efficiently. Firstly, an adaptive parameter adjustment mechanism is designed, based on population diversity and fitness. This enables better-informed adaptation of related parameters of genetic operators. Secondly, the mutation operator guided by the gene potential contribution of genes tends to produce better offspring. The gene potential contribution reflects the positive effect of the current gene on fitness. It guides the evolution by weighting the more valuable genes with the distribution index matrix. Finally, the concept of aging is introduced into the environmental selection, to offer more opportunities to the young generation and alleviate premature convergence. The proposed algorithm has been evaluated on eight different datasets, and compared against 44 state-of-the-art algorithms from the literature. The experimental results show that the network designed by AG-ENAS obtains higher classification accuracy than manual-designed networks such as SENet, DenseNet, and other ENAS algorithms such as Large-Evo and AE-CNN. [ABSTRACT FROM AUTHOR]

Background Cardiac surgery is associated with a significant decrease in hematocrit. It is unclear whether that occurs from hemodilution, loss of red cells, or both. Hematocrit is a major determinant of transfusion decisions although transfusion is associated with increased morbidity and mortality. Physicians must determine whether this anemia is the result of hemodilution or red blood cell loss as the former would be treated with packed red blood cell transfusions and the latter by diuresis. We hypothesize that the decrease in hematocrit observed in cardiac surgery is due to hemodilution. Methods Blood volume (BV), plasma volume (PV), and red blood cell volume (RBCV) were measured in 54 patients undergoing coronary artery bypass graft surgery, valve surgery, or coronary artery bypass graft/valve surgery. Measurements were made preoperatively, immediately postoperatively, and 2 hours after surgery utilizing a dilution tracer method and hematocrit measurements. Results Preoperative average BV was 6,094 mL (SD 1,904 mL), RBCV was 2,024 mL (SD 720 mL), and PV was 4,070 mL (SD 1,339 mL). Postoperative average BV was 4,834 mL (SD 1,432 mL), RBCV 1,226 mL (SD 527 mL), and PV 3,607 mL (SD 993 mL). Blood volume decreased 18% ( p < 0.0001), RBCV decreased 38% ( p < 0.001), and PV decreased 8% ( p < 0.012). There were no significant changes between postoperative values and those 2 hours later in the cardiac surgery intensive care unit. Conclusions Decreases in hematocrit observed in cardiac surgery patients are due to significant red blood cell losses and not to hemodilution. Red blood cell losses averaged 38%. Plasma volume also decreased. [ABSTRACT FROM AUTHOR]