Generative Learning in VLSI Design for Manufacturability: Current Status and Future Directions

With the continuous scaling of integrated circuit technologies,design for manufacturability(DFM)is becoming more critical,yet more challenging.Alongside,recent advances in machine learning have provided a new computing paradigm with promising applications in VLSI manufacturability.In particular,generative learning-regarded among the most interesting ideas in present-day machine learning-has demonstrated impressive capabilities in a wide range of applications.This paper surveys recent results of using generative learning in VLSI manufacturing modeling and optimization.Specifically,we examine the unique features of generative learning that have been leveraged to improve DFM efficiency in an unprecedented way;hence,paving the way to a new data-driven DFM approach.The state-of-the-art methods are presented,and challenges/opportunities are discussed.

A Dual Discriminator Method for Generalized Zero-Shot Learning

Zero-shot learning enables the recognition of new class samples by migrating models learned from semanticfeatures and existing sample features to things that have never been seen before. The problems of consistencyof different types of features and domain shift problems are two of the critical issues in zero-shot learning. Toaddress both of these issues, this paper proposes a new modeling structure. The traditional approach mappedsemantic features and visual features into the same feature space;based on this, a dual discriminator approachis used in the proposed model. This dual discriminator approach can further enhance the consistency betweensemantic and visual features. At the same time, this approach can also align unseen class semantic features andtraining set samples, providing a portion of information about the unseen classes. In addition, a new feature fusionmethod is proposed in the model. This method is equivalent to adding perturbation to the seen class features,which can reduce the degree to which the classification results in the model are biased towards the seen classes.At the same time, this feature fusion method can provide part of the information of the unseen classes, improvingits classification accuracy in generalized zero-shot learning and reducing domain bias. The proposed method isvalidated and compared with othermethods on four datasets, and fromthe experimental results, it can be seen thatthe method proposed in this paper achieves promising results.

Generative Adversarial Networks for Secure Data Transmission in Wireless Network

In this paper,a communication model in cognitive radios is developed and uses machine learning to learn the dynamics of jamming attacks in cognitive radios.It is designed further to make their transmission decision that automati-cally adapts to the transmission dynamics to mitigate the launched jamming attacks.The generative adversarial learning neural network(GALNN)or genera-tive dynamic neural network(GDNN)automatically learns with the synthesized training data(training)with a generator and discriminator type neural networks that encompass minimax game theory.The elimination of the jamming attack is carried out with the assistance of the defense strategies and with an increased detection rate in the generative adversarial network(GAN).The GDNN with game theory is designed to validate the channel condition with the cross entropy loss function and back-propagation algorithm,which improves the communica-tion reliability in the network.The simulation is conducted in NS2.34 tool against several performance metrics to reduce the misdetection rate and false alarm rates.The results show that the GDNN obtains an increased rate of successful transmis-sion by taking optimal actions to act as a defense mechanism to mislead the jam-mer,where the jammer makes high misclassification errors on transmission dynamics.

Solar image deconvolution by generative adversarial network

With aperture synthesis(AS)technique,a number of small antennas can be assembled to form a large telescope whose spatial resolution is determined by the distance of two farthest antennas instead of the diameter of a single-dish antenna.In contrast from a direct imaging system,an AS telescope captures the Fourier coefficients of a spatial object,and then implement inverse Fourier transform to reconstruct the spatial image.Due to the limited number of antennas,the Fourier coefficients are extremely sparse in practice,resulting in a very blurry image.To remove/reduce blur,“CLEAN”deconvolution has been widely used in the literature.However,it was initially designed for a point source.For an extended source,like the Sun,its efficiency is unsatisfactory.In this study,a deep neural network,referring to Generative Adversarial Network(GAN),is proposed for solar image deconvolution.The experimental results demonstrate that the proposed model is markedly better than traditional CLEAN on solar images.The main purpose of this work is visual inspection instead of quantitative scientific computation.We believe that this will also help scientists to better understand solar phenomena with high quality images.

GACS:Generative Adversarial Imitation Learning Based on Control Sharing

Generative adversarial imitation learning(GAIL)directly imitates the behavior of experts from human demonstration instead of designing explicit reward signals like reinforcement learning.Meanwhile,GAIL overcomes the defects of traditional imitation learning by using a generative adversary network framework and shows excellent performance in many fields.However,GAIL directly acts on immediate rewards,a feature that is reflected in the value function after a period of accumulation.Thus,when faced with complex practical problems,the learning efficiency of GAIL is often extremely low and the policy may be slow to learn.One way to solve this problem is to directly guide the action(policy)in the agents'learning process,such as the control sharing(CS)method.This paper combines reinforcement learning and imitation learning and proposes a novel GAIL framework called generative adversarial imitation learning based on control sharing policy(GACS).GACS learns model constraints from expert samples and uses adversarial networks to guide learning directly.The actions are produced by adversarial networks and are used to optimize the policy and effectively improve learning efficiency.Experiments in the autonomous driving environment and the real-time strategy game breakout show that GACS has better generalization capabilities,more efficient imitation of the behavior of experts,and can learn better policies relative to other frameworks.

Conditional Generative Adversarial Networks for modelling fuel sprays

In this study, the probabilistic, data driven nature of the generative adversarial neural networks (GANs)was utilized to conduct virtual spray simulations for conditions relevant to aero engine combustors. Themodel consists of two sub-modules: (i) an autoencoder converting the variable length droplet trajectories intofixed length, lower dimensional representations and (ii) a Wasserstein GAN that learns to mimic the latentrepresentations of the evaporating droplets along their lifetime. The GAN module was also conditioned withthe injection location and the diameters of the droplets to increase the generalizability of the whole framework.The training data was provided from highly resolved 3D, transient Eulerian–Lagrangian, large eddy simulationsconducted with OpenFOAM. Neural network models were created and trained within the open source machinelearning framework of PyTorch. Predictive capabilities of the proposed method was discussed with respect tospray statistics and evaporation dynamics. Results show that conditioned GAN models offer a great potentialas low order model approximations with high computational efficiency. Nonetheless, the capabilities of theautoencoder module to preserve local dependencies should be improved to realize this potential. For the currentcase study, the custom model architecture was capable of conducting the simulation in the order of secondsafter a day of training, which had taken one week on HPC with the conventional CFD approach for the samenumber of droplets (200,000 trajectories).

Research on Generalized Computing Systems

This paper presents a kind of artificial intelligent system-generalized computing system (GCS for short), and introduces its mathematical description, implement problem and learning problem.

Comparative study of low NO_(x) combustion optimization of a coal-fired utility boiler based on OBLPSO and GOBLPSO

To reduce NO_(x) emissions of coal-fired power plant boilers,this study introduced particle swarm optimization employing opposition-based learning(OBLPSO)and particle swarm optimization employing generalized opposition-based learning(GOBLPSO)to a low NO_(x) combustion optimization area.Thermal adjustment tests under different ground conditions,variable oxygen conditions,variable operation modes of coal pulverizer conditions,and variable first air pressure conditions were carried out on a 660 MW boiler to obtain samples of combustion optimization.The adaptability of PSO,differential evolution algorithm(DE),OBLPSO,and GOBLPSO was compared and analyzed.Results of 51 times independently optimized experiments show that PSO is better than DE,while the performance of the GOBLPSO algorithm is generally better than that of the PSO and OBLPSO.The median-optimized NO_(x) emission by GOBLPSO is up to 15.8 mg/m^(3) lower than that obtained by PSO.The generalized opposition-based learning can effectively utilize the information of the current search space and enhance the adaptability of PSO to the low NO_(x) combustion optimization of the studied boiler.

A kind of Generalized Learning Model

This paper presents a kind of new machine learning model——generalized learning model, with brain adaptive principles as its theory foundation and generalized learning units as its basic elements. Several generalized learning units may constitute a learning network. All kind of sense analysis networks, integration networks and memory networks form generalized learning system. The generalized learning system possesses many characteristics such as high parallel, effective fusion and optimized evolution.

A novel collaborative decision-making method based on generalized abductive learning for resolving design conflicts

In complex product design,lots of time and resources are consumed to choose a preference-based compromise decision from non-inferior preliminary design models with multi-objective conflicts.However,since complex products involve intensive multi-domain knowledge,preference is not only a comprehensive representation of objective data and subjective knowledge but also characterized by fuzzy and uncertain.In recent years,enormous challenges are involved in the design process,within the increasing complexity of preference.This article mainly proposes a novel decision-making method based on generalized abductive learning(G-ABL)to achieve autonomous and efficient decision-making driven by data and knowledge collaboratively.The proposed G-ABL framework,containing three cores:classifier,abductive kernel,and abductive machine,supports preference integration from data and fuzzy knowledge.In particular,a subtle improvement is presented for WK-means based on the entropy weight method(EWM)to address the local static weight problem caused by the fixed data preferences as the decision set is locally invariant.Furthermore,fuzzy comprehensive evaluation(FCE)and Pearson correlation are adopted to quantify domain knowledge and obtain abducted labels.Multi-objective weighted calculations are utilized only to label and compare solutions in the final decision set.Finally,an engineering application is provided to verify the effectiveness of the proposed method,and the superiority of which is illustrated by comparative analysis.

GLCM Based Extraction of Flame Image Texture Features and KPCA-GLVQ Recognition Method for Rotary Kiln Combustion Working Conditions

According to the pulverized coal combustion flame image texture features of the rotary-kiln oxide pellets sintering process,a combustion working condition recognition method based on the generalized learning vector(GLVQ) neural network is proposed.Firstly,the numerical flame image is analyzed to extract texture features,such as energy,entropy and inertia,based on grey-level co-occurrence matrix(GLCM) to provide qualitative information on the changes in the visual appearance of the flame.Then the kernel principal component analysis(KPCA) method is adopted to deduct the input vector with high dimensionality so as to reduce the GLVQ target dimension and network scale greatly.Finally,the GLVQ neural network is trained by using the normalized texture feature data.The test results show that the proposed KPCA-GLVQ classifer has an excellent performance on training speed and correct recognition rate,and it meets the requirement for real-time combustion working condition recognition for the rotary kiln process.

Generalized Oppositional Moth Flame Optimization with Crossover Strategy:An Approach for Medical Diagnosis

In the original Moth-Flame Optimization(MFO),the search behavior of the moth depends on the corresponding flame and the interaction between the moth and its corresponding flame,so it will get stuck in the local optimum easily when facing the multi-dimensional and high-dimensional optimization problems.Therefore,in this work,a generalized oppositional MFO with crossover strategy,named GCMFO,is presented to overcome the mentioned defects.In the proposed GCMFO,GOBL is employed to increase the population diversity and expand the search range in the initialization and iteration jump phase based on the jump rate;crisscross search(CC)is adopted to promote the exploitation and/or exploration ability of MFO.The proposed algorithm’s performance is estimated by organizing a series of experiments;firstly,the CEC2017 benchmark set is adopted to evaluate the performance of GCMFO in tackling high-dimensional and multimodal problems.Secondly,GCMFO is applied to handle multilevel thresholding image segmentation problems.At last,GCMFO is integrated into kernel extreme learning machine classifier to deal with three medical diagnosis cases,including the appendicitis diagnosis,overweight statuses diagnosis,and thyroid cancer diagnosis.Experimental results and discussions show that the proposed approach outperforms the original MFO and other state-of-the-art algorithms on both convergence speed and accuracy.It also indicates that the presented GCMFO has a promising potential for application.