Gradient descent, computed through backpropagation (BP), has been widely used to train spiking neural networks (SNNs). However, the approach has several limitations. It requires manual intervention to tune the network architecture, is prone to catastrophic forgetting of previously learned information when exposed to data containing new information, and is computationally demanding. To address these issues, we propose brain-mimetic developmental spiking neural networks (BDNNs), which emulate the postnatal development of biological neural circuits. We evaluated BDNNs using a neuromorphic tactile system with the task of classifying objects through grasping. Our findings show that BDNNs grow dynamically in response to input data by incrementally recruiting hidden neurons, leading to steadily increasing classification accuracy without the need for manual architecture tuning. The growth process adapts autonomously to the complexity of incoming data. BDNNs also exhibit strong knowledge transfer capabilities, which effectively leverage previously learned knowledge about grasping objects to incrementally learn about new objects. Furthermore, in comparative experiments using the same dataset and hardware, BDNNs achieved classification performance comparable to the standard BP-based method and its variants, while learning one to three orders of magnitude faster. Furthermore, the BDNN outperforms existing continual learning algorithms in the performance and speed. These results highlight BDNNs as a promising approach for continual learning and real-time edge computing applications. The source code of our work is publicly available at https://github.com/1jiaqixing/BDNNversion1.

Recent progress in novel view synthesis of indoor scenes using diffusion models has attracted significant attention, particularly for generating desired poses from a source image. Existing methods can produce plausible views near the input view, but they often fail to extrapolate views far beyond the input perspective. Additionally, achieving a multiview consistent diffusion model typically requires training of computationally intensive 3D priors, limiting scalability to long-range generation. In this paper, we present a transformer-based latent diffusion model that leverages view geometry constraints, including explicitly warped feature maps of the input view as the denoised target view and a conditioning combination of epipolar-weighted source image feature map, Plücker raymap, and camera poses. This approach allows for the extrapolation of consistent novel views, both semantically and geometrically, over long-range trajectories in a single-shot manner. Our model is evaluated on two indoor datasets, ScanNet and RealEstate10K, using a diverse set of metrics for view quality and consistency evaluation. Experimental results demonstrate the superiority of our approach over existing models, showcasing its potential for semantically and geometrically consistent novel view synthesis, scalable in video generation applications.

Novel view synthesis (NVS) from a single image is highly ill-posed due to large unobserved regions, especially for views that deviate significantly from the input. While existing methods focus on consistency between the source and generated views, they often fail to maintain coherence and correct view alignment across long-range or looped trajectories. We propose a model that addresses this by decomposing single-view NVS into a 360-degree scene extrapolation followed by novel view interpolation. This design ensures long-term view and scene consistency by conditioning on keyframes extracted and warped from a generated panoramic representation. In the first stage, a panorama diffusion model learns the scene prior from the input perspective image. Perspective keyframes are then sampled and warped from the panorama and used as anchor frames in a pre-trained video diffusion model, which generates novel views through a proposed spatial noise diffusion process. Compared to the prior work, our method produces globally consistent novel views-even in loop-closure scenarios, while enabling flexible camera control. Experiments on diverse scene datasets demonstrate that our approach outperforms existing methods in generating coherent views along user-defined trajectories. Our implementation is available at https://github.com/YiGuYT/LookBeyond.

Few-shot Named Entity Recognition aims to identify named entities from unstructured texts in various domains using a minimal amount of training samples and classify them into predefined categories. Many popular approaches decompose this process into two tasks: Span Detection and Entity Classification. However, they still have some issues: (1) Neglecting presentation optimization. Most of these methods emphasize classification, neglecting the optimization of span presentation during Span Detection. (2) Missing label semantics. They have not fully leveraged semantic information of entity type labels during Entity Classification. To address these issues, this paper proposes Multi-Granularity Dual-Aware Contrastive Learning (MGDAC) for few-shot NER. Specifically, we introduce multi-granularity contrastive learning for solve neglecting presentation optimization, focusing on both the overall vector and internal vector granularity to enhance features beneficial for span detection in token vector representations. Additionally, we design dual-aware contrastive learning for solve missing label semantics, effectively utilizing semantic information from both entity tokens and entity type labels during prototype construction to jointly optimize prototype representations. Finally, extensive experiments on the FewNERD dataset demonstrate that our proposed method exhibits improvements in both Span Detection and Entity Classification, outperforming other competitive baseline methods.

Recent years has witnessed the remarkable success in end-to-end task-oriented dialog system, especially when incorporating external knowledge information.However, the quality of most existing models' generated response is still limited, mainly due to their lack of finegrained reasoning on deterministic knowledge (w.r.t.conceptual tokens), which makes them difficult to capture the concept shifts and identify user's real intention in cross-task scenarios.To address these issues, we propose a novel intention mechanism to better model deterministic entity knowledge.Based on such a mechanism, we further propose an intention reasoning network (IR-Net), which consists of joint and multi-hop reasoning, to obtain intention-aware representations of conceptual tokens that can be used to capture the concept shifts involved in task-oriented conversations, so as to effectively identify user's intention and generate more accurate responses.Experimental results verify the effectiveness of IR-Net, showing that it achieves the stateof-the-art performance on two representative multi-domain dialog datasets.

Foreign exchange is the largest financial market in the world, and it is also one of the most volatile markets. Technical analysis plays an important role in the forex market and trading algorithms are designed utilizing machine learning techniques. Most literature used historical price information and technical indicators for training. However, the noisy nature of the market affects the consistency and profitability of the algorithms. To address this problem, we designed trading rule features that are derived from technical indicators and trading rules. The parameters of technical indicators are optimized to maximize trading performance. We also proposed a novel cost function that computes the risk-adjusted return, Sharpe and Sterling Ratio (SSR), in an effort to reduce the variance and the magnitude of drawdowns. An automatic robotic trading (RoboTrading) strategy is designed with the proposed Genetic Algorithm Maximizing Sharpe and Sterling Ratio model (GA-MSSR) model. The experiment was conducted on intraday data of 6 major currency pairs from 2018 to 2019. The results consistently showed significant positive returns and the performance of the trading system is superior using the optimized rule-based features. The highest return obtained was 320% annually using 5-minute AUDUSD currency pair. Besides, the proposed model achieves the best performance on risk factors, including maximum drawdowns and variance in return, comparing to benchmark models. The code can be accessed at https://github.com/zzzac/rule-based-forextrading-system.