Search

Deep generative models applied to the generation of novel compounds in small-molecule drug design have attracted a lot of attention in recent years. To design compounds that interact with specific target proteins, we propose a Generative Pre-Trained Transformer (GPT)-inspired model for de novo target-specific molecular design. By implementing different keys and values for the multi-head attention conditional on a specified target, the proposed method can generate drug-like compounds both with and without a specific target. The results show that our approach (cMolGPT) is capable of generating SMILES strings that correspond to both drug-like and active compounds. Moreover, the compounds generated from the conditional model closely match the chemical space of real target-specific molecules and cover a significant portion of novel compounds. Thus, the proposed Conditional Generative Pre-Trained Transformer (cMolGPT) is a valuable tool for de novo molecule design and has the potential to accelerate the molecular optimization cycle time. [ABSTRACT FROM AUTHOR]

[Display omitted] • GANs newly used for topological design of metaporous materials for sound absorption. • Huge acceleration (~0.04s/design) for design process enabling instantaneous designing. • Successful designs of broadband sound absorption checked by simulation and experiment. • Creative configurations and rich local features generated in GANs-designed patterns. • AI-guided designing/optimizing as new possibility for AI-materials interdiscipline. The topological design and optimization of metaporous materials is one of the key challenges in the field of sound absorption. Limited by the expensive computational cost, it is particularly disadvantaged when instantaneous multiple designs are required. In recent years, an increasing number of research fields are harnessing machine learning approaches thanks to their experience-free manner and outstanding efficiency. Generative Adversarial Networks (GANs), as a type of machine learning algorithms, enjoy the special benefit of powerful generative capability, making them brilliantly suitable for designing purposes. Additionally, it can fully explore the data distribution space with enormous computational power and create brand new designs. In this work, GANs are newly employed for the topological design of metaporous materials for sound absorption. Trained with numerically prepared data, they successfully propose designs with high-standard broadband absorption performance, verified by simulation and experiment. The designing process is dramatically accelerated by hundreds of times using GANs (100 designs in 4.372 s). This allows GANs to easily provide more structures and configurations, and achieve instantaneous multiple solutions, giving designers more choices to satisfy various constraints such as mass or porosity. In addition, GANs are demonstrated remarkably capable of generating creative configurations and rich local features. This work proposes a new designing principle, illustrates the value of machine learning in guiding the designing and optimizing process in the mechanical world, and opens new possibilities for the future of AI-materials interdisciplinary research. [ABSTRACT FROM AUTHOR]

• Predict the sound absorption curve from 300 Hz to 3000 Hz (50 Hz interval) with one single deep learning network. • Build residual connection network blocks to obtain network models of different depths. • Determine the most suitable network hyperparameter via constantly monitoring the overfitting level of all the models. • Exploit cross-validation to train the network to the best performance. Obtaining the airborne sound absorption coefficient is essential for studying the sound absorption performance and sound absorption mechanism of acoustic metamaterials. The most commonly used method for numerical calculation of airborne sound absorption coefficient is Finite Element Method (FEM). However, when the number of samples is relatively large, especially when the internal geometric structure of the samples is complicated, the calculation cost of FEM becomes exponentially high. Compared with FEM, machine learning algorithms show great potential in efficiently and intelligently predicting material properties. Taking images representing the topological structure of acoustic metamaterials (along with their airborne sound absorption performance simulated by FEM) as input, we propose a deep convolutional neural network to predict the broadband airborne sound absorption curve of the metaporous materials from 300 Hz to 3000 Hz with the interval of 50 Hz. To avoid overfitting, the network hyperparameter with favorable generalization capability is determined via constantly monitoring the overfitting level of the network. In addition, cross-validation is exploited to train the network to the best performance. Designed in such a compact manner where only one network is sufficient to predict for a whole absorption curve with a large range, the network is marvelously computationally economic and efficient and shows excellent prediction accuracy. [ABSTRACT FROM AUTHOR]

[Display omitted] • SAP-net with CNN to predict the sound absorption coefficient of metaporous materials. • Enormous acceleration achieved (7ms/image) for the prediction process. • Trained SAP-nets show outstanding accuracy with MAE averagely smaller than 0.02. • SAP-net's ability to learn the underlying mechanism of topology-performance linkage. • New fast and accurate tool for investigating sound absorption and designing materials. Airborne sound absorption coefficient is the premise for investigating the sound absorption performance or mechanism of metaporous materials. The common numerical evaluation approach is FEM which is relatively computationally costly particularly when processing complex structures or a large batch of data. Rapidly developing deep learning algorithms, on the other hand, show a promising trend in the data-driven manner to learn and predict material parameters efficiently and precisely. We propose SAP-net based on deep convolutional neural network to predict the sound absorption coefficient at a specific frequency of an input image representing the topological structure of metaporous materials. Trained with FEM-prepared data for six frequency points, SAP-net demonstrates outstanding evaluation speed of 0.007 s/image and brilliant prediction accuracy with mean absolute errors all smaller than 0.019 (the smallest 0.008 at f = 1000 Hz). Meanwhile, the fact that SAP-net remains accurate when predicting for images that are essentially different from those in the training data shows its capability of learning and capturing the underlying physical mechanism linking the topological structure to the sound absorption performance. In conclusion, SAP-net provides an extraordinarily fast and accurate approach for the investigation of sound absorption performance, which is expected to accelerate the examination and design process of materials. [ABSTRACT FROM AUTHOR]