Neuromemetic Evolutionary Optimization

The design of a new therapeutic agent is a time-consuming and expensive process. The rise of machine intelligence provides a grand opportunity of expeditiously discovering novel drug candidates through smart search in the vast molecular structural space. In this paper, we propose a new approach called adversarial deep evolutionary learning (ADEL) to search for novel molecules in the latent space of an adversarial generative model and keep improving the latent representation space. In ADEL, a custom-made adversarial autoencoder (AAE) model is developed and trained under a deep evolutionary learning (DEL) process. This involves an initial training of the AAE model, followed by an integration of multi-objective evolutionary optimization in the continuous latent representation space of the AAE rather than the discrete structural space of molecules. By using the AAE, an arbitrary distribution can be provided to the training of AAE such that the latent representation space is set to that distribution. This allows for a starting latent space from which new samples can be produced. Throughout the process of learning, new samples of high-quality are generated after each iteration of training and then added back into the full dataset. Therefore, allowing for a more comprehensive procedure of understanding the data structure. This combination of evolving data and continuous learning not only enables improvement in the generative model, but the data as well. By comparing ADEL to the previous work in DEL, we see that ADEL can obtain better property distributions.

NRC-VABS: Normalized Reparameterized Conditional Variational Autoencoder with applied beam search in latent space for drug molecule design

The hardware design space is high-dimensional and discrete. Systematic and efficient exploration of this space has been a significant challenge. Central to this problem is the intractable search complexity that grows exponentially with the design choices and the discrete nature of the search space. This work investigates the feasibility of learning a meaningful low-dimensional continuous representation for hardware designs to reduce such complexity and facilitate the search process. We devise a variational autoencoder (VAE)-based design space exploration framework called VAESA, to encode the hardware design space in a compact and continuous representation. We show that black-box and gradient-based design space exploration algorithms can be applied to the latent space, and design points optimized in the latent space can be reconstructed to high-performance realistic hardware designs. Our experiments show that performing the design space search on the latent space consistently leads to the optimal design point under a fixed number of samples. In addition, the latent space can improve the sample efficiency of the original algorithm by 6.8$\times$ and can discover hardware designs that are up to 5% more efficient than the optimal design searched directly in the high-dimensional input space.

A deep Koopman operator‐based modelling approach for long‐term prediction of dynamics with pixel‐level measurements

A person’s gait is a behavioral trait that is uniquely associated with each individual and can be used to recognize the person. As information about the human gait can be captured by wearable devices, a few studies have led to the proposal of methods to process gait information for identification purposes. Despite recent advances in gait recognition, an open set gait recognition problem presents challenges to current approaches. To address the open set gait recognition problem, a system should be able to deal with unseen subjects who have not included in the training dataset. In this paper, we propose a system that learns a mapping from a multimodal time series collected using insole to a latent (embedding vector) space to address the open set gait recognition problem. The distance between two embedding vectors in the latent space corresponds to the similarity between two multimodal time series. Using the characteristics of the human gait pattern, multimodal time series are sliced into unit steps. The system maps unit steps to embedding vectors using an ensemble consisting of a convolutional neural network and a recurrent neural network. To recognize each individual, the system learns a decision function using a one-class support vector machine from a few embedding vectors of the person in the latent space, then the system determines whether an unknown unit step is recognized as belonging to a known individual. Our experiments demonstrate that the proposed framework recognizes individuals with high accuracy regardless they have been registered or not. If we could have an environment in which all people would be wearing the insole, the framework would be used for user verification widely.

Configuration selection is an important and often primary task of conceptual design teams within aerospace. Many other sub-tasks within the domain have heavily benefited and today even rely upon computer-aided design methods, but configuration selection has remained a largely abstract, informal process. In this paper we develop a model of configuration selection as a NP-complete mixed-integer optimization problem to explicate the difficulties involved therein and introduce EDEN - the Experimental Design Engineering Network - as a class of generative neural networks whose inference may be used as a heuristic method for identifying promising design candidates. We demonstrate methods for embedding design candidates into a continuous latent space and building a synthetic, performance-weighted probability distribution over said space. By learning to reconstruct samples generated from this latent space as design candidates, EDEN surpasses previous work in generative design of configurations by predictably generating high-quality design candidates rather than merely feasible configurations with no weighting toward optimality.

Recent works on domain adaptation reveal the effectiveness of adversarial learning on filling the discrepancy between source and target domains. However, two common limitations exist in current adversarial-learning-based methods. First, samples from two domains alone are not sufficient to ensure domain-invariance at most part of latent space. Second, the domain discriminator involved in these methods can only judge real or fake with the guidance of hard label, while it is more reasonable to use soft scores to evaluate the generated images or features, i.e., to fully utilize the inter-domain information. In this paper, we present adversarial domain adaptation with domain mixup (DM-ADA), which guarantees domain-invariance in a more continuous latent space and guides the domain discriminator in judging samples' difference relative to source and target domains. Domain mixup is jointly conducted on pixel and feature level to improve the robustness of models. Extensive experiments prove that the proposed approach can achieve superior performance on tasks with various degrees of domain shift and data complexity.

Human mobility clustering is an important problem for understanding human mobility behaviors (e.g., work and school commutes). Existing methods typically contain two steps: choosing/learning a mobility representation and applying a clustering algorithm to the representation. However, these methods rely on strict visiting orders in trajectories and cannot take advantage of multiple types of mobility representations. This paper proposes a novel mobility clustering method for mobility behavior detection. First, the proposed method contains a permutation-equivalent operation to handle sub-trajectories that might have different visiting orders but similar impacts on mobility behaviors. Second, the proposed method utilizes a variational autoencoder architecture to simultaneously perform clustering in both latent and original spaces. Also, in order to handle the bias of a single latent space, our clustering assignment prediction considers multiple learned latent spaces at different epochs. This way, the proposed method produces accurate results and can provide reliability estimates of each trajectory’s cluster assignment. The experiment shows that the proposed method outperformed state-of-the-art methods in mobility behavior detection from trajectories with better accuracy and more interpretability.

We consider multi-solution optimization and generative models for the generation of diverse artifacts and the discovery of novel solutions. In cases where the domain's factors of variation are unknown or too complex to encode manually, generative models can provide a learned latent space to approximate these factors. When used as a search space, however, the range and diversity of possible outputs are limited to the expressivity and generative capabilities of the learned model. We compare the output diversity of a quality diversity evolutionary search performed in two different search spaces: 1) a predefined parameterized space and 2) the latent space of a variational autoencoder model. We find that the search on an explicit parametric encoding creates more diverse artifact sets than searching the latent space. A learned model is better at interpolating between known data points than at extrapolating or expanding towards unseen examples. We recommend using a generative model's latent space primarily to measure similarity between artifacts rather than for search and generation. Whenever a parametric encoding is obtainable, it should be preferred over a learned representation as it produces a higher diversity of solutions.

Molecular optimization plays a pivotal role in many domains since it holds promise for improving the properties of lead molecules. The advent of artificial intelligence (AI)-driven molecular optimization has revolutionized lead optimization workflows, which have significantly accelerated the development of drug candidates. However, AI models are also confronted with new challenges in practical molecular optimization, such as high-dimensional chemical space and data sparsity issues. This paper initially highlights the inherent benefits of molecular optimization in terms of optimizing the properties and maintaining the structural similarity of lead molecules, thereby highlighting its critical role in drug discovery. The next section systematically categorizes and analyzes existing AI-aided molecular optimization methods, comprising iterative search in discrete chemical space, end-to-end generation in continuous latent space, and iterative search in continuous latent space methods. Finally, we discuss the key challenges in AI-aided molecular optimization methods, including molecular representations, dataset selection, the properties to be optimized, and optimization algorithms, while proposing potential solutions and future research directions. In summary, this review provides a comprehensive analysis of existing representative AI-aided molecular optimization methods, thereby offering guidance for future research directions.

Physics‐aware deep learning (PADL) has gained popularity for use in spatiotemporal dynamics simulations, such as those in computational modeling of energetic materials (EM). We show that the challenge PADL methods face while learning complex field evolution problems can be simplified and accelerated by decoupling it into two tasks: learning complex geometric features in evolving fields and modeling dynamics over these features in a lower‐dimensional feature space. We build upon our previous work on physics‐aware recurrent convolutional neural networks (PARC). PARC embeds knowledge of underlying physics into its neural network architecture for more robust and accurate prediction of evolving physical fields. PARC was shown to effectively learn complex nonlinear features such as the formation of hotspots and coupled shock fronts in various initiation scenarios of EMs, as a function of microstructures, serving effectively as a microstructure‐aware burn model. Here, we further accelerate PARC and reduce its computational cost by projecting the original dynamics onto a lower‐dimensional invariant manifold, or “latent space.” The projected latent representation encodes the complex geometry of evolving fields (e.g., temperature and pressure) in a set of data‐driven features. The reduced dimension of this latent space allows us to learn the dynamics during the initiation of EM with a lighter and more efficient model. We observe a significant decrease in training and inference time while maintaining results comparable to PARC at inference. This work takes steps towards enabling rapid prediction of EM thermomechanics at larger scales and characterization of EM structure–property–performance linkages at a full application scale.

In optimization and machine learning, the divide between discrete and continuous problems and methods is deep and persistent. We attempt to remove this distinction by training neural network autoencoders that embed discrete candidate solutions in continuous latent spaces. This allows us to take advantage of state-of-the-art continuous optimization methods for solving discrete optimization problems, and mitigates certain challenges in discrete optimization, such as design of bias-free search operators. In the experimental part, we consider program synthesis as the special case of combinatorial optimization. We train an autoencoder network on a large sample of programs in a problem-agnostic, unsupervised manner, and then use it with an evolutionary continuous optimization algorithm (CMA-ES) to map the points from the latent space to programs. We propose also a variant in which semantically similar programs are more likely to have similar embeddings. Assessment on a range of benchmarks in two domains indicates the viability of this approach and the usefulness of involving program semantics.

High dimensional pose state space is the main challenge in articulated human pose tracking which makes pose analysis computationally expensive or even infeasible. In this paper, we propose a novel generative approach in the framework of evolutionary computation, by which we try to widen the bottleneck with effective search strategy embedded in the extracted state subspace. Firstly, we use ISOMAP to learn the low-dimensional latent space of pose state in the aim of both reducing dimensionality and extracting the prior knowledge of human motion simultaneously. Then, we propose a manifold reconstruction method to establish smooth mappings between the latent space and original space, which enables us to perform pose analysis in the latent space. In the search strategy, we adopt a new evolutionary approach, clonal selection algorithm (CSA), for pose optimization. We design a CSA based method to estimate human pose from static image, which can be used for initialization of motion tracking. In order to make CSA suitable for motion tracking, we propose a sequential CSA (S-CSA) algorithm by incorporating the temporal continuity information into the traditional CSA. Actually, in a Bayesian inference view, the sequential CSA algorithm is in essence a multilayer importance sampling based particle filter. Our methods are demonstrated in different motion types and different image sequences. Experimental results show that our CSA based pose estimation method can achieve viewpoint invariant 3D pose reconstruction and the S-CSA based motion tracking method can achieve accurate and stable tracking of 3D human motion.

Model Inversion (MI) attacks based on Generative Adversarial Networks (GAN) aim to recover private training data from complex deep learning models by searching codes in the latent space. However, this method merely searches in a deterministic latent space, resulting in suboptimal latent codes. Additionally, existing distributional MI schemes assume that an attacker can access the structures and parameters of the target model, which is not always feasible in practice. To address these limitations, this paper proposes a novel Distributional Black-Box Model Inversion (DBB-MI) attack by constructing a probabilistic latent space for searching private data. Specifically, DBB-MI does not require the target model’s parameters or specialized GAN training. Instead, it identifies the latent probability distribution by integrating the output of the target model with multi-agent reinforcement learning techniques. Then, it randomly selects latent codes from the latent probability distribution to uncover private data. As the latent probability distribution closely mirrors the target privacy data in the latent space, the recovered data effectively leaks the privacy of the target model’s training samples. Extensive experiments conducted on diverse datasets and networks demonstrate that our DBB-MI outperforms state-of-the-art MI attacks in terms of attack accuracy, K-nearest neighbor feature distance, and peak signal-to-noise ratio.

Summary Full-waveform inversion (FWI) can capture high-resolution geophysical properties but tends to fall into local minima. We addressed this problem by reducing the degree of freedom in the optimization problem and estimating an initial velocity model from shot gathers; our method comprises three steps. First, we trained a variational autoencoder using synthetic velocity models and created a low-dimensional latent space that represents realistic velocity models. Second, we prepared another encoder model, which estimates the latent features of the corresponding velocity model from shot gathers. Finally, we implemented latent space optimization (LSO) from the initial guess selected by the encoder. In the experiment, LSO reproduced much more accurate and realistic velocity models than the naive FWI, even when the initial guess was considerably different from the ground truth. This implies that with our method the latent space becomes easier to optimize than the original velocity model space. Furthermore, we verified that LSO using an initial guess selected by the encoder could increase the estimation accuracy. Our method also succeeded in accurately estimating the Marmousi2 model, without using any part of the benchmark dataset to directly train the variational autoencoder.