Implicit Neural Representations (INRs) have become a powerful paradigm for modeling continuous signals in computer vision, graphics, and scientific computing. However, multilayer perceptrons (MLPs) generally suffer from severe spectral bias, which limits their ability to accurately model high-frequency details and multi-scale structures. To address this challenge, we propose a novel Multi-Scale Sine Activation (MSA), which explicitly introduces multi-scale frequency responses by incorporating multiple sets of sine activations with logarithmically spaced frequencies in parallel at each layer. MSA is further combined with an amplitude modulation mechanism to ensure numerical stability and robust optimization across different frequency channels. We conduct extensive experiments on a series of challenging tasks, including 1D multi-scale function fitting, image representation, video representation, 3D shape representation, and PDEs solving. Experimental results show that MSA outperforms existing state-of-the-art methods in terms of reconstruction accuracy, detail preservation, and training stability.

The quest for analytical solutions to differential equations has traditionally been constrained by the need for extensive mathematical expertise. Machine learning methods like genetic algorithms have shown promise in this domain, but are hindered by significant computational time and the complexity of their derived solutions. This paper introduces **SSDE** (Symbolic Solver for Differential Equations), a novel reinforcement learning-based approach that derives symbolic closed-form solutions for various differential equations. Evaluations across a diverse set of ordinary and partial differential equations demonstrate that SSDE outperforms existing machine learning methods, delivering superior accuracy and efficiency in obtaining analytical solutions.

Exposure problems often affect image quality more significantly than noise and blurring, seriously affecting their effectiveness in computer vision applications. To address this core problem, this paper proposes an innovative deep-learning method. The method uses pyramidal multi-scale decomposition to repair the image layer by layer, handling both overexposure and underexposure problems. Firstly, we introduce the Strengthen-Operate-Subtract (SOS) feature enhancement module, which refines and enhances the image by building previously estimated images to make the exposure correction effect more pronounced. Additionally, we propose using the Multi-scale Aggregation Back-projection (MAB) module to improve the U-Net model. Through the error feedback mechanism, the features of non-adjacent layers are compensated and integrated effectively, and the loss of detailed information is reduced. We introduce a dense block of matte residuals at the maximum scale to expand the network's receptive field and effectively capture local information. These innovations significantly enhance image exposure correction, reducing detail loss, regional contrast saturation, and color imbalance. Extensive evaluations confirm the effectiveness of our model, which demonstrates superior performance across all datasets, particularly in PSNR, achieving an average value of 26.271.

Summary Image salient region detection methods, which can detect and extract interesting regions in pictures, have been a hot research direction in recent years. Most of current salient region detection algorithms and corresponding training datasets originated from general visual attention processes and primarily reflected attention to object shape in pictures. While color vision provided more useful information in vision systems, the color effect of visual attention should be and must be considered. First, we collected cue data of painting pictures from several observers using eye‐track recording technology when observers were asked to pay attention to color information of various paintings. Second, we constructed a color attention dataset, color saliency dataset (CSD), from the cue data and pictures. Thirdly, we designed a V‐fused color saliency net (VCSNet) model which included three modules: a color information fusion module, a prediction module, and an optimization module and trained the model using the CSD. Finally, we compared our method with previous algorithms on the CSD, and results showed that our method outperformed the previous algorithms in color saliency detection with MAE of 0.057 and Fmax of 0.265. We open source part of the self‐created dataset: https://github.com/InfiniteEM/ColorSaliencyDataSet .

The incremental learning paradigm in machine learning has consistently been a focus of academic research. It is similar to the way in which biological systems learn, and reduces energy consumption by avoiding excessive retraining. Existing studies utilize the powerful feature extraction capabilities of pre-trained models to address incremental learning, but there remains a problem of insufficient utilization of neural network feature knowledge. To address this issue, this paper proposes a novel method called Pre-trained Model Knowledge Distillation (PMKD) which combines knowledge distillation of neural network representations and replay. This paper designs a loss function based on centered kernel alignment to transfer neural network representations knowledge from the pre-trained model to the incremental model layer-by-layer. Additionally, the use of memory buffer for Dark Experience Replay helps the model retain past knowledge better. Experiments show that PMKD achieved superior performance on various datasets and different buffer sizes. Compared to other methods, our class incremental learning accuracy reached the best performance. The open-source code is published at https://github.com/TianSongS/PMKD-IL .

This paper describes a special-purpose neural computing system for face identification. The system architecture and hardware implementation are introduced in detail. An algorithm based on biomimetic pattern recognition has been embedded. For the total 1200 tests for face identification, the false rejection rate is 3.7% and the false acceptance rate is 0.7%.