The control of the compressor in fuel cell systems (FCS) plays a crucial role in ensuring stable and efficient operation. To further enhance the efficiency and durability of FCS, considering the impact of power alloca...
详细信息
Describing images with natural and fluent sentences is recognized as a challenging problem in computer vision and natural language processing. The limited number of studies in this area, especially in the Turkish lang...
详细信息
The remarkable generalization of large-scale models has recently gained significant attention in multimodal ***, deploying heterogeneous large-scale models with different modalities under Federated Learning (FL) to pr...
详细信息
The remarkable generalization of large-scale models has recently gained significant attention in multimodal ***, deploying heterogeneous large-scale models with different modalities under Federated Learning (FL) to protect data privacy imposes tremendous challenges on clients' limited computation and *** this work, we propose M2FEDSA to address the above *** realize modularized decomposition of large-scale models via Split Learning (SL) and only retain privacy-sensitive modules on clients, alleviating storage *** freezing large-scale models and introducing two specialized lightweight adapters, the models can better focus on task-specific knowledge and enhance modality-specific knowledge, improving the model's adaptability to different tasks while balancing *** addition, M2FEDSA further improves performance by transferring multimodal knowledge to unimodal clients at both the feature and decision levels, which leverages the complementarity of different *** experiments on various multimodal classification tasks validate the effectiveness of our proposed *** code is made available publicly at https://***/M2FedSA/M-2FedSA. Copyright 2024 by the author(s)
In the supervisory control layer of cyber-physical systems, attackers may introduce delays and disorder into the communication channel used for transmitting commands between the supervisor and the controlled system, t...
详细信息
This paper proposes a cascaded high gain DC-DC converter based on the switched capacitor and coupled inductor with merits of high voltage-boost capability, low voltage stress and only one active switch. The combinatio...
详细信息
Data collection and annotation are time-consuming in machine learning, especially for large-scale problems. A com-mon approach for these problems is to transfer knowledge from the related labeled domain to the target ...
详细信息
Periodic B-spline (PBS) has been a successfully used technique in histopathology image segmentation. However, it is limited when dealing with multi-instance objects in real-world biomedical applications. Moreover, its...
详细信息
The perception of salty taste is crucial for individuals to make healthy food ***,the brain electrophysiological signals underlying salty taste perception have been poorly *** this study,electroencephalography(EEG)was...
详细信息
The perception of salty taste is crucial for individuals to make healthy food ***,the brain electrophysiological signals underlying salty taste perception have been poorly *** this study,electroencephalography(EEG)was used to record brain activity induced by Na Cl solution as a salty taste stimulus.A combination of a custom delivery device and stimulation paradigm was employed to preserve the salty taste signal clearly.A stimulus-response capture method was proposed that could adapt to individual differences in brain responses to salty taste and accurately segment salty taste response *** this method to the EEG processing workflow can form a complete data processing *** results showed that the neural response induced by salty taste reached a high activity level in the initial stage within a short period(0.2 s),and there was a sustained periodic response within 0.75 s after the ***,the salty taste information in the EEG signal was decoded,and discrimination of 2 similar concentrations of salty taste solutions was achieved far above the chance level(average identification rate:89.66%).This study demonstrated experimental paradigms and research methods for understanding salty taste perception,which could provide references for research on other basic tastes.
Relative overgeneralization (RO) occurs in cooperative multi-agent learning tasks when agents converge towards a suboptimal joint policy due to overfitting to suboptimal behaviors of other *** methods have been propos...
详细信息
Relative overgeneralization (RO) occurs in cooperative multi-agent learning tasks when agents converge towards a suboptimal joint policy due to overfitting to suboptimal behaviors of other *** methods have been proposed for addressing RO in multi-agent policy gradient (MAPG) methods although these methods produce state-of-the-art *** address this gap, we propose a general, yet simple, framework to enable optimistic updates in MAPG methods that alleviate the RO *** approach involves clipping the advantage to eliminate negative values, thereby facilitating optimistic updates in *** optimism prevents individual agents from quickly converging to a local ***, we provide a formal analysis to show that the proposed method retains optimality at a fixed *** extensive evaluations on a diverse set of tasks including the Multi-agent MuJoCo and Overcooked benchmarks, our method outperforms strong baselines on 13 out of 19 tested tasks and matches the performance on the rest. Copyright 2024 by the author(s)
Aiming at the problem of speed fluctuation of permanent magnet synchronous motor (PMSM) caused by periodic load change during the operation of scroll air compressor, a sensorless control method based on linear active ...
详细信息
暂无评论