Conditional Adversarial Autoencoder Research Articles

Risk-Conditioned Reinforcement Learning: A Generalized Approach for Adapting to Varying Risk Measures

In application domains requiring mission-critical decision making, such as finance and robotics, the optimal policy derived by reinforcement learning (RL) often hinges on a preference for risk management. Yet, the dynamic nature of risk measures poses considerable challenges to achieving generalization and adaptation of risk-sensitive policies in the context of RL. In this paper, we propose a risk-conditioned RL model that enables rapid policy adaptation to varying risk measures via a unified risk representation, the Weighted Value-at-Risk (WV@R). To sample risk measures that avoid undue optimism, we construct a risk proposal network employing a conditional adversarial auto-encoder and a normalizing flow. This network establishes coherent representations for risk measures, preserving the continuity in terms of the Wasserstein distance on the risk measures. The normalizing flow is used to support non-crossing quantile regression that obtains valid samples for risk measures, and it is also applied to the agent’s critic to ascertain the preservation of monotonicity in quantile estimations. Through experiments with locomotion, finance, and self-driving scenarios, we show that our model is capable of adapting to a range of risk measures, achieving comparable performance to the baseline models individually trained for each measure. Our model often outperforms the baselines, especially in the cases when exploration is required during training but risk-aversion is favored during evaluation.

Investigation the influence related to parameters configuration of Generative Adversarial Networks in face image generation

Due to the excellent performance of Generative Adversarial Networks (GAN) for age regression on face images, it is particularly important to explore the effect of different parameters on model training. In this study, the origin and development of Artificial Intelligence (AI) is first discussed, from which the concept and principles of GAN are derived. This is followed by a brief introduction of the UTKface dataset used in this research, and the Conditional Adversarial Autoencoder (CAAE) framework based on the GAN technique. The division of labor and roles of the encoder, generator, and the two discriminators in the model are described. The various learning rates as well as batch size combinations attempted in this study are then illustrated, and the training results of the model are shown in the form of graphs and plots of the loss value function. A situation where the model stops learning is highlighted in the results, which is similar to pattern descent in GAN, and is shown to be characterized by the inability of the discriminator to successfully recognize it. Ultimately, drawing from the acquired outcomes, it can be deduced that employing a larger batch size serves to enhance the pace of model training. It is advisable to concurrently elevate the learning rate by an equivalent factor when augmenting the batch size, thereby ensuring a consistent trajectory for model convergence.

Face Aging by Explainable Conditional Adversarial Autoencoders

This paper deals with Generative Adversarial Networks (GANs) applied to face aging. An explainable face aging framework is proposed that builds on a well-known face aging approach, namely the Conditional Adversarial Autoencoder (CAAE). The proposed framework, namely, xAI-CAAE, couples CAAE with explainable Artificial Intelligence (xAI) methods, such as Saliency maps or Shapley additive explanations, to provide corrective feedback from the discriminator to the generator. xAI-guided training aims to supplement this feedback with explanations that provide a "reason" for the discriminator's decision. Moreover, Local Interpretable Model-agnostic Explanations (LIME) are leveraged to provide explanations for the face areas that most influence the decision of a pre-trained age classifier. To the best of our knowledge, xAI methods are utilized in the context of face aging for the first time. A thorough qualitative and quantitative evaluation demonstrates that the incorporation of the xAI systems contributed significantly to the generation of more realistic age-progressed and regressed images.

Cross-age face synthesis based on conditional adversarial autoencoder

Face aging aims to render face images with desired age attribute. It has tremendous impact to a wide-range of applications, e.g., criminal investigation, entertainment. The rapid development of generative adversarial networks (GANs) has shown impressive results in face aging. Among them, the Conditional Adversarial Autoencoder (CAAE) proposed in 2017 has achieved good results in face aging. However, the generated faces still have the problems that the aging features are not obvious and the identity information are not well maintained. In addition, research have shown that the human aging process is affected by genes. Different races have different external characteristics of aging. However, the current research does not take the racial factor into account, ignores the racial differences in the aging process. It affects the accuracy of transformation. To solve the above problems, this paper proposes a cross-age face synthesis based on conditional adversarial autoencoder: First, a conditional adversarial autoencoder is used as the infrastructure to build a cross-age face synthesis model based on race constraints. Secondly, the discriminator is composed of a discriminant network and a classification network, and a category loss function is designed to generate a real face that matches the target age; Finally, the model uses a identity feature extractor and a discriminator of the multi-scale architecture. Through multi-level discrimination from pixel values to high-level semantic information, the loss of personal identity features is minimized. UTKFace and MegaAge-Asian datasets are used in the experiment. Three comparative experiments are designed for the above improvements. The results show that the racial constraints make the generated images effectively maintain the racial characteristics, such as skin color and texture; The classification function of the discriminator improves the aging effect; The design of the multi-scale discriminator enables the generated face to have more stable local structure and identity information. Through qualitative and quantitative analysis, it is shown that this method has higher aging accuracy and identity retention than the CAAE method.

Face Age Feature Analysis Based on Improved Conditional Adversarial Auto-encoder (I-CAAE)

<p>In recent years, the research of face age features has achieved rapid development driven by deep learning. The faces generated by the Conditional Adversarial Auto-encoder (CAAE) model are not only highly credible, but also closer to the target age. However, there are many problems, such as low resolution of human face image generation and poor local feature retention effect of human face features. To this end, this paper improves on the CAAE network. Firstly, referring to the LSGAN network structure, the 4 convolution layers of the encoder are added to 5 layers and the 4 convolution layers of the generator are added to 7 layers. Secondly, on the basis of the original loss function, the image gradient difference loss function is added to ensure the output face image quality. Meanwhile, the data set were preprocessed for face correction. Finally, this paper performs face similarity analysis on the Eye-key platform and contrasts the generated image quality using structural similarity and peak signal to noise ratio metrics. In addition, the generated results were tested for their robustness. The experimental results show that the average similarity of faces generated by the Improved Conditional Adversarial Auto-encoder (I-CAAE) network was increased by 3.9. And the average peak signal to noise ratio of the generated pictures was reduced by 1.8. Confirming the superiority of the proposed method.</p> <p> </p>

Development of a Face Prediction System for Missing Children in a Smart City Safety Network

Cases of missing children not being found are rare, but they continue to occur. If the child is not found immediately, the parents may not be able to identify the child’s appearance because they have not seen their child for a long time. Therefore, our purpose is to predict children’s faces when they grow up and help parents search for missing children. DNA paternity testing is the most accurate way to detect whether two people have a blood relation. However, DNA paternity testing for every unidentified child would be costly. Therefore, we propose the development of the Face Prediction System for Missing Children in a Smart City Safety Network. It can predict the faces of missing children at their current age, and parents can quickly confirm the possibility of blood relations with any unidentified child. The advantage is that it can eliminate incorrect matches and narrow down the search at a low cost. Our system combines StyleGAN2 and FaceNet methods to achieve prediction. StyleGAN2 is used to style mix two face images. FaceNet is used to compare the similarity of two face images. Experiments show that the similarity between predicted and expected results is more than 75%. This means that the system can well predict children’s faces when they grow up. Our system has more natural and higher similarity comparison results than Conditional Adversarial Autoencoder (CAAE), High Resolution Face Age Editing (HRFAE) and Identity-Preserved Conditional Generative Adversarial Networks (IPCGAN).

Distilling from professors: Enhancing the knowledge distillation of teachers

Knowledge distillation (KD) is a successful technique for transferring knowledge from one machine learning model to another model. Specifically, the idea of KD has been widely used for various tasks such as model compression and knowledge transfer between different models. However, existing studies in KD have overlooked the possibility that dark knowledge (i.e., soft targets) obtained from a complex and large model (a.k.a., a teacher model) may be either incorrect or insufficient. Such knowledge can hinder the effective learning of another small model (a.k.a., a student model). In this paper, we propose the professor model, which refines the soft target from the teacher model to improve KD. The professor model aims to achieve two goals; 1) improving the prediction accuracy and 2) capturing the inter-class correlation of the soft target from the teacher model. We first design the professor model by reformulating a conditional adversarial autoencoder (CAAE). Then, we devise two KD strategies using both teacher and professor models. Our empirical study demonstrates that the professor model effectively improves KD in three benchmark datasets: CIFAR100, TinyImagenet, and ILSVRC2015. Moreover, our comprehensive analysis shows that the professor model is much more effective than employing the stronger teacher model, in which parameters are greater than the sum of the teacher’s and professor’s parameters. Since the proposed model is model-agnostic, our model can be combined with any KD algorithm and consistently improves various KD techniques.

Affective Latent Representation of Acoustic and Lexical Features for Emotion Recognition.

In this paper, we propose a novel emotion recognition method based on the underlying emotional characteristics extracted from a conditional adversarial auto-encoder (CAAE), in which both acoustic and lexical features are used as inputs. The acoustic features are generated by calculating statistical functionals of low-level descriptors and by a deep neural network (DNN). These acoustic features are concatenated with three types of lexical features extracted from the text, which are a sparse representation, a distributed representation, and an affective lexicon-based dimensions. Two-dimensional latent representations similar to vectors in the valence-arousal space are obtained by a CAAE, which can be directly mapped into the emotional classes without the need for a sophisticated classifier. In contrast to the previous attempt to a CAAE using only acoustic features, the proposed approach could enhance the performance of the emotion recognition because combined acoustic and lexical features provide enough discriminant power. Experimental results on the Interactive Emotional Dyadic Motion Capture (IEMOCAP) corpus showed that our method outperformed the previously reported best results on the same corpus, achieving 76.72% in the unweighted average recall.

Face image synthesis with weight and age progression using conditional adversarial autoencoder

The appearance of a human face changes with the change in body weight and age. With varying lifestyle choices, it is hard to imagine the appearance of a given human face in years to come. Future self-perception is highly associated with one’s emotional state, as well as health behavior. Negative future self-perception can cause negative lifestyle choice and negative health behavior, leading to depression and eating disorder. In this paper, a new methodology is introduced for future self-face image synthesis using age and weight, resulting in visualization of future face image derived from given weight category and age. A Constrained Local Model is first used for weight progressed future face image synthesized and then age-progressed future face image is generated using Conditional Adversarial Auto Encoder. In the final step, both weight progressed and age-progressed face images fed to face morphing module which synthesized future face image by keeping natural looks. Experimental results show the advantages of proposed method with promising results.

Entangled Conditional Adversarial Autoencoder for de Novo Drug Discovery.

Modern computational approaches and machine learning techniques accelerate the invention of new drugs. Generative models can discover novel molecular structures within hours, while conventional drug discovery pipelines require months of work. In this article, we propose a new generative architecture, entangled conditional adversarial autoencoder, that generates molecular structures based on various properties, such as activity against a specific protein, solubility, or ease of synthesis. We apply the proposed model to generate a novel inhibitor of Janus kinase 3, implicated in rheumatoid arthritis, psoriasis, and vitiligo. The discovered molecule was tested in vitro and showed good activity and selectivity.

CAAE++: Improved CAAE for Age Progression/Regression

Face age progression/regression has garnered substantial active research interest due to its tremendous impact on a wide-range of practical applications like searching for missing individuals with photos of childhood, entertainment, and so on. Most existing face aging models have proven to be successful and effective in learning the transformation between age groups with the aid of paired samples, i.e., face images of the same person at different ages. Considering the expensive cost of collecting paired datasets, Conditional Adversarial Autoencoder (CAAE) is designed for face aging task without paired samples and first achieves face age progression and regression in a holistic framework. However, only rough wrinkles are generated because of the insufficient discriminative and generative ability. To tackle this problem, in this paper, we develop a novel generative model based on CAAE, dubbed CAAE++, which defeats the previous CAAE mainly for two enhancements: 1) an auxiliary classifier is added on top of the discriminator, which allows a single discriminator not only distinguishes real images from synthetics but also classifies them into the target age group; and 2) a pre-trained deep face recognition model and a pre-trained age estimation model are exploited to preserve identity and age similarity, respectively. We train CAAE++ on UTKFace dataset and test on FGNET dataset. Experimental results demonstrate the efficacy of our proposed method in terms of fidelity.

Face Aging With Boundary Equilibrium Conditional Autoencoder

Since generative adversarial networks (GANs) were proposed in 2014, mode collapse has been a problem that affects many researchers when training GANs. With the reconstruction loss of an autoencoder, conditional adversarial autoencoder (CAAE) is free from mode collapse. However, its reconstruction loss will bring a saturation problem, in which the encoder maps every input image into just one latent variable. Combining the CAAE with a boundary equilibrium generative adversarial network, we propose a boundary equilibrium conditional autoencoder (BECAE) focusing on the face aging task. Our model is the first GANs that renders images through a discriminator. We also introduce some statistics to measure the level of the saturation problem. The results show that the BECAE has successfully solved the saturation problem and can generate face images of the same quality as the images generated by the CAAE.