We propose a novel and versatile computational approach, based on hierarchical COSFIRE filters, that addresses the challenge of explainable retina and palmprint recognition for automatic person identification. Unlike ...
详细信息
We propose a novel and versatile computational approach, based on hierarchical COSFIRE filters, that addresses the challenge of explainable retina and palmprint recognition for automatic person identification. Unlike traditional systems that treat these biometrics separately, our method offers a unified solution, leveraging COSFIRE filters’ trainable nature for enhanced selectivity and robustness, while exhibiting explainability and resilience to decision-based black-box adversarial attack and partial matching. COSFIRE filters are trainable, in that their selectivity can be determined with a one-shot learning step. In practice, we configure a COSFIRE filter that is selective for the mutual spatial arrangement of a set of automatically selected keypoints of each retina or palmprint reference image. A query image is then processed by all COSFIRE filters and it is classified with the reference image that was used to configure the COSFIRE filter that gives the strongest similarity score. Our approach, tested on the VARIA and RIDB retina datasets and the IITD palmprint dataset, achieved state-of-the-art results, including perfect classification for retina datasets and a 97.54% accuracy for the palmprint dataset. It proved robust in partial matching tests, achieving over 94% accuracy with 80% image visibility and over 97% with 90% visibility, demonstrating effectiveness with incomplete biometric data. Furthermore, while effectively resisting a decision-based black-box adversarial attack and impervious to imperceptible adversarial images, it is only susceptible to highly perceptible adversarial images with severe noise, which pose minimal concern as they can be easily detected through histogram analysis in preprocessing. In principle, the proposed learning-free hierarchical COSFIRE filters are applicable to any application that requires the identification of certain spatial arrangements of moderately complex features, such as bifurcations and crossovers. Moreover, the sele
Enhancing the dialogue interactions of Non-Player Characters (NPCs) in video games, this work introduces a novel integration of Large Language Models (LLMs) with Pursuit Learning Automata (PLA). The approach is design...
详细信息
ISBN:
(数字)9798350394634
ISBN:
(纸本)9798350394641
Enhancing the dialogue interactions of Non-Player Characters (NPCs) in video games, this work introduces a novel integration of Large Language Models (LLMs) with Pursuit Learning Automata (PLA). The approach is designed to foster dynamic, engaging, and contextually relevant conversations within a highly resource-efficient framework. Utilizing the generative strengths of LLMs alongside the adaptive learning properties of PLA, the system presented here dynamically modulates dialogue tones and emotions. This ensures a tailored gaming experience that does not require online LLM processing. Initial findings suggest that this method boosts player engagement and satisfaction, contributing to the development of more immersive and responsive gaming worlds.
A flotation circuit is controlled in simulation using an extremum seeking control (ESC) approach to keep the cells operating at the optimal operating point, as represented by peak air recovery. It is assumed that opti...
详细信息
An optimization computational grid algorithm and the results of its application for estimating the velocity characteristics of complex medium based on experimentally obtained seismic wave arrival times are presented. ...
详细信息
The Internet is booming with information, and it has become especially difficult for consumers to sift through the information. Recommendation systems can effectively enhance the consumer experience. However, model-ba...
详细信息
The Internet is booming with information, and it has become especially difficult for consumers to sift through the information. Recommendation systems can effectively enhance the consumer experience. However, model-based recommendation systems require sufficient training data, so they perform poorly in small-scale recommendation scenarios such as graduate school recommendation. To this end, we focus on online recommendation in graduate school application scenarios. We propose a Pre-purify Temporal-decay Memory-based Collaborative Filtering model called PTMCF, which firstly improves the data quality based on the users’ background information by pre-purifying the data to compensate for the poor performance caused by the small dataset. At the same time, considering that user preferences and the importance of information are constantly changing, we propose incorporating Newton’s Law of Cooling when constructing the user-item scoring matrix to assign time-based weights. Experiments on a dataset collected from real-world questionnaires show that pre-purify and temporal-decay effectively improve recommendation quality and mitigate the impact of data sparsity on memory-based collaborative filtering. IEEE
The use of Double Fed Induction Machines (DFIM) is widely implemented in wind farms and pumping stations due to their regulation capacity. The rotor of DFIMs is one of the most expensive parts of these installation. T...
详细信息
Reinforcement Learning (RL) has garnered much attention in the field of control due to its capacity to learn from interactions and adapt to complex and dynamic environments. However, RL is challenging because it needs...
详细信息
ISBN:
(数字)9798350330991
ISBN:
(纸本)9798350331004
Reinforcement Learning (RL) has garnered much attention in the field of control due to its capacity to learn from interactions and adapt to complex and dynamic environments. However, RL is challenging because it needs to balance exploration, seeking new strategies, and exploitation, leveraging known strategies for maximum gain. To address these challenges, this paper proposes a Model Predictive Control (MPC) based RL approach, where the state value function in RL is utilized as the cost function in MPC, and the system dynamic model is represented by neural networks (NNs). This eliminates the need for human intervention and addresses inaccuracies in the system model. Additionally, MPC-guided RL accelerates convergence during RL training, thereby enhancing sample efficiency. Reported results demonstrate that the proposed method outperforms traditional RL algorithms and does not require prior knowledge of the system.
In modern society, deaf and mute individuals face significant challenges in social and daily communication due to their limitations in speech expression. Although traditional assistive tools have made some progress in...
详细信息
In this paper, a bistable material based on glass fiber reinforced resin matrix composites (GFRP) is used as the dielectric substrate of a printed dipole antenna to realize pattern reconfigurable characteristics. The ...
详细信息
Real networks are complex dynamical systems, evolving over time with the addition and deletion of nodes and links. Currently, there exists no principled mathematical theory for their dynamics—a grand-challenge open p...
详细信息
暂无评论