Deep Evolutionary Learning for Molecular Design

In this paper, a prototypical deep evolutionary learning (DEL) process is proposed to integrate deep generative model and multi-objective evolutionary computation for molecular design. Our approach enables (1) evolutionary operations in the latent space of the generative model, rather than the structural space, to generate promising novel molecular structures for the next evolutionary generation, and (2) generative model fine-tuning using newly generated high-quality samples. Thus, DEL implements a data-model co-evolution concept which improves both sample population and generative model learning. Experiments on public datasets indicate that the sample population obtained by DEL exhibits improvement on property distributions, and dominates samples generated by other baseline molecular optimization algorithms. Furthermore, comparisons with a range of deep generative models show that DEL is beneficial for improving sample populations. Read More

IEEE Computational Intelligence Magazine, May 2022

Deep Learning in Physics: a Study of Dielectric Quasi-Cubic Particles in a Uniform Electric Field

Solving physics problems for which we know the equations, boundary conditions and symmetries can be done by deep learning. Here, we calculate the induced field inside and outside a dielectric quasi-cube placed in a uniform electric field, wherein the dielectric mismatch at edges and corners of the cube makes accurate calculations challenging. The electric potential is expressed as an ansatz incorporating neural networks with known leading order forms and symmetries, then Laplace's equation with boundary conditions at the dielectric interface is solved by minimizing a loss function inside a large solution domain. We study how the electric potential inside and outside the particle evolves through a sequence of shapes from a sphere to a cube. The neural network being differentiable, it is straightforward to calculate the electric field over the whole domain, the induced surface charge distribution and the polarizability. The neural network being retentive, one can efficiently follow how the field changes upon particle's shape or dielectric constant by iterating from previously converged solution.  Read More

IEEE Transactions on Emerging Topics in Computational Intelligence, June 2022

Constructing ANFIS With Sparse Data Through Group-Based Rule Interpolation: An Evolutionary Approach

An adaptive-network-based fuzzy inference system (ANFIS) offers a popular and powerful fuzzy inference mechanism. As with many other advanced data-driven techniques, developing an effective ANFIS typically requires sufficient training data. However, in many real-world applications, it is not always straightforward to obtain a large amount of representative data that cover the entire problem space to accomplish the required training, seriously restricting the performance of a learned ANFIS. This article introduces a new ANFIS learning approach through an evolutionary process, which is able to generate an ANFIS with only a small amount of training data in a certain problem region, by interpolating well-trained ANFISs in the neighboring regions. Such a process works by first producing an initial population of candidate fuzzy rules in the region of data shortage, through interpolating a rule dictionary constructed from trained ANFISs in the neighborhood regions.  Read more

IEEE Transactions on Fuzzy Systems, April 2022

In Search of a Neural Model for Serial Order: A Brain Theory for Memory Development and Higher Level Cognition

In order to keep trace of information and grow up, the infant brain has to resolve the problem about where old information is located and how to index new ones. We propose that the immature prefrontal cortex (PFC) uses its primary functionality of detecting hierarchical patterns in temporal signals as a second feature to organize the spatial ordering of the cortical networks in the developing brain itself. Our hypothesis is that the PFC detects the hierarchical structure in temporal sequences in the shape of ordinal patterns and uses them to index information hierarchically in different parts of the brain. Henceforth, we propose that this mechanism for detecting ordinal patterns participates also in the hierarchical organization of the brain during development; i.e., the bootstrapping of the connectome. Read More

IEEE Transactions on Cognitive and Developmental Systems, June 2022

Anomaly Detection on Attributed Networks via Contrastive Self-Supervised Learning

Anomaly detection on attributed networks attracts considerable research interests due to wide applications of attributed networks in modeling a wide range of complex systems. Recently, the deep learning-based anomaly detection methods have shown promising results over shallow approaches, especially on networks with high-dimensional attributes and complex structures. However, existing approaches, which employ graph autoencoder as their backbone, do not fully exploit the rich information of the network, resulting in suboptimal performance. Furthermore, these methods do not directly target anomaly detection in their learning objective and fail to scale to large networks due to the full graph training mechanism. To overcome these limitations, in this article, we present a novel Contrastive self-supervised Learning framework for Anomaly detection on attributed networks (CoLA for abbreviation). Our framework fully exploits the local information from network data by sampling a novel type of contrastive instance pair, which can capture the relationship between each node and its neighboring substructure in an unsupervised way. Read more

IEEE Transactions on Neural Networks and Learning Systems, June 2022

Designing Game Feel: A Survey

Game feel design is the intentional design of the affective impact of moment-to-moment interaction with games. In this article, we survey academic research and publications by practitioners to give a complete overview of the state of research concerning this aspect of game design. We analyzed over 200 sources and categorized their content according to design purposes. This resulted in three different domains of intended player experiences: 1) physicality; 2) amplification; and 3) support. In these domains, the act of polishing, which determines game feel, takes the shape of tuning, juicing, and streamlining, respectively. Tuning the physicality of game objects creates cohesion and predictability, and the resulting movement informs level design. Juicing is the act of polishing amplification and it results in empowerment and provides clarity of feedback. Read more

IEEE Transactions on Games, June 2022