Smart farming, also known as precision agriculture or digital farming, is an innovative approach to agriculture that utilizes advanced technologies and data-driven techniques to optimize various aspects of farming ope...
详细信息
Testing model transformations poses several challenges, one of which is how to automatically generate effective test suites. A promising approach for this is to employ equivalence partitioning, a well-known technique ...
详细信息
Audio Deepfakes, which are highly realistic fake audio recordings driven by AI tools that clone human voices, With Advancements in Text-Based Speech Generation (TTS) and Vocal Conversion (VC) technologies have enabled...
详细信息
Audio Deepfakes, which are highly realistic fake audio recordings driven by AI tools that clone human voices, With Advancements in Text-Based Speech Generation (TTS) and Vocal Conversion (VC) technologies have enabled it easier to create realistic synthetic and imitative speech, making audio Deepfakes a common and potentially dangerous form of deception. Well-known people, like politicians and celebrities, are often targeted. They get tricked into saying controversial things in fake recordings, causing trouble on social media. Even kids’ voices are cloned to scam parents into ransom payments, etc. Therefore, developing effective algorithms to distinguish Deepfake audio from real audio is critical to preventing such frauds. Various Machine learning (ML) and Deep learning (DL) techniques have been created to identify audio Deepfakes. However, most of these solutions are trained on datasets in English, Portuguese, French, and Spanish, expressing concerns regarding their correctness for other languages. The main goal of the research presented in this paper is to evaluate the effectiveness of deep learning neural networks in detecting audio Deepfakes in the Urdu language. Since there’s no suitable dataset of Urdu audio available for this purpose, we created our own dataset (URFV) utilizing both genuine and fake audio recordings. The Urdu Original/real audio recordings were gathered from random youtube podcasts and generated as Deepfake audios using the RVC model. Our dataset has three versions with clips of 5, 10, and 15 seconds. We have built various deep learning neural networks like (RNN+LSTM, CNN+attention, TCN, CNN+RNN) to detect Deepfake audio made through imitation or synthetic techniques. The proposed approach extracts Mel-Frequency-Cepstral-Coefficients (MFCC) features from the audios in the dataset. When tested and evaluated, Our models’ accuracy across datasets was noteworthy. 97.78% (5s), 98.89% (10s), and 98.33% (15s) were remarkable results for the RNN+LSTM
Reinforcement learning(RL) has been widely adopted for intelligent decision-making in embodied agents due to its effective trial-and-error learning capabilities. However, most RL methods overlook the causal relationsh...
详细信息
Reinforcement learning(RL) has been widely adopted for intelligent decision-making in embodied agents due to its effective trial-and-error learning capabilities. However, most RL methods overlook the causal relationships among states and actions during policy exploration and lack the human-like ability to distinguish signal from noise and reason with important abstractions, resulting in poor sample efficiency. To address this issue, we propose a novel method named causal action empowerment(CAE) for efficient RL, designed to improve sample efficiency in policy learning for embodied agents. CAE identifies and leverages causal relationships among states, actions, and rewards to extract controllable state variables and reweight actions for prioritizing high-impact behaviors. Moreover, by integrating a causality-aware empowerment term, CAE significantly enhances an embodied agent's execution of causally-aware behavior for more efficient exploration via boosting controllability in complex embodied environments. Benefiting from these two improvements, CAE bridges the gap between local causal discovery and global causal empowerment. To comprehensively evaluate the effectiveness of CAE, we conduct extensive experiments across 25 tasks in 5 diverse embodied environments, encompassing both locomotion and manipulation skill learning with dense and sparse reward settings. Experimental results demonstrate that CAE consistently outperforms existing methods across this wide range of scenarios, offering a promising avenue for improving sample efficiency in RL.
In-situ fabricated polyether electrolytes have been regarded as one of the most promising solid electrolyte systems. Nevertheless, they cannot match high-voltage cathodes over 4.3 V due to their poor oxidative stabili...
详细信息
In-situ fabricated polyether electrolytes have been regarded as one of the most promising solid electrolyte systems. Nevertheless, they cannot match high-voltage cathodes over 4.3 V due to their poor oxidative stability. Herein, we propose an effective local charge homogenization strategy based on the triglycidyl isocyanurate（TGIC） crosslinker, achieving ultra-high-voltage electrochemical stability of polyether electrolytes（viz. PTIDOL） at cutoff voltages up to 4.7 V. The introduction of TGIC optimizes the Li+solvation environment, thereby homogenizing the charge distribution at ether oxygen（EO） sites, resulting in significantly enhanced oxidative stability of the polyether main chain. Consequently, the Li|PTIDOL|LiNi0.6Co0.2Mn0.2O2（NCM622） cell achieves long-term operation at an ultra-high cutoff voltage with a capacity retention of 81.8% after 400 cycles, one of the best results reported for polyether electrolytes to date. This work provides significant insights for the development of polyether electrolytes with high-voltage tolerance and the advancement of high-energy-density batteries.
Artificial Intelligence (AI) is transforming numerous domains, including bioinformatics and information extraction systems, by advancing data processing capabilities, enhancing precision, and facilitating automation. ...
详细信息
Sampling and communication are both crucial for coordination in multi-agent systems(MASs), with sampling capturing raw data from the environment for control decision making, and communication ensuring the data is shar...
详细信息
Sampling and communication are both crucial for coordination in multi-agent systems(MASs), with sampling capturing raw data from the environment for control decision making, and communication ensuring the data is shared effectively for synchronized and informed control decisions across agents. However, practical MASs often operate in environments where continuous and synchronous data samplings and exchanges are impractical, necessitating strategies that can handle intermittent sampling and communication constraints. This paper provides a comprehensive survey of recent advances in distributed coordination control of MASs under intermittent sampling and communication, focusing on both foundational principles and state-of-the-art techniques. After introducing fundamentals, such as communication topologies,agent dynamics, control laws, and typical coordination objectives, the distinctions between sampling and communication are elaborated, exploring deterministic versus random, synchronous versus asynchronous, and instantaneous versus sequential scenarios. A detailed review of emerging trends and techniques is then presented, covering time-triggered, event-triggered,communication-protocol-based, and denial-of-service-resilient coordination control. These techniques are analyzed across various attack models, including those based on data loss, sampled data, time constraints, and topology switching. By synthesizing these developments, this survey aims to equip researchers and practitioners with a clearer understanding of current challenges and methodologies, concluding with insights into promising future directions.
All-vanadium flow batteries（VFBs） are one of the most promising large-scale energy storage *** an operando quantitative analysis of the polarizations in VFBs under different conditions is essential for developing hi...
详细信息
All-vanadium flow batteries（VFBs） are one of the most promising large-scale energy storage *** an operando quantitative analysis of the polarizations in VFBs under different conditions is essential for developing high power density ***,we employ an operando decoupling method to quantitatively analyze the polarizations in each electrochemical and chemical reaction of VFBs under different catalytic *** show that the reduction reaction of V3+presents the largest activation polarization,while the reduction reaction of VO2+primarily contributes to concentration polarizations due to the formation of the intermediate product V2O3+.Additionally,it is found that the widely used electrode catalytic methods,incorporating oxygen functional groups and electrodepositing Bi,not only enhance the reaction kinetics but also exacerbate concentration polarizations simultaneously,especially during the discharge ***,in the battery with the high oxygen-containing electrodes,the negative side still accounts for the majority of activation loss（75.3%） at 200 mA cm-2,but it comes down to 36,9% after catalyzing the negative reactions with *** work provides an effective way to probe the limiting steps in flow batteries under various working conditions and offers insights for effectively enhancing battery performance for future developments.
This paper introduces a novel lightweight colour image encryption algorithm,specifically designed for resource-constrained environments such as Internet of Things(IoT)*** IoT systems become increasingly prevalent,secu...
详细信息
This paper introduces a novel lightweight colour image encryption algorithm,specifically designed for resource-constrained environments such as Internet of Things(IoT)*** IoT systems become increasingly prevalent,secure and efficient data transmission becomes *** proposed algorithm addresses this need by offering a robust yet resource-efficient solution for image *** image encryption relies on confusion and diffusion *** stages are generally implemented linearly,but this work introduces a new RSP(Random Strip Peeling)algorithm for the confusion step,which disrupts linearity in the lightweight category by using two different sequences generated by the 1D Tent Map with varying initial *** diffusion stage then employs an XOR matrix generated by the Logistic *** evaluation metrics,such as entropy analysis,key sensitivity,statistical and differential attacks resistance,and robustness analysis demonstrate the proposed algorithm's lightweight,robust,and *** proposed encryption scheme achieved average metric values of 99.6056 for NPCR,33.4397 for UACI,and 7.9914 for information entropy in the SIPI image *** also exhibits a time complexity of O(2×M×N)for an image of size M×N.
Dear Editor,This letter presents a new transfer learning framework for the deep multi-agent reinforcement learning(DMARL) to reduce the convergence difficulty and training time when applying DMARL to a new scenario [1...
详细信息
Dear Editor,This letter presents a new transfer learning framework for the deep multi-agent reinforcement learning(DMARL) to reduce the convergence difficulty and training time when applying DMARL to a new scenario [1], [2].
暂无评论