Globally, the greatest concerns of farmers are plant diseases. A considerable loss of yield has an immediate impact on the economy. Machine learning models exhibited capabilities to detect plant diseases. To enhance t...
详细信息
The management of plastic waste in our ecosystem represents a significant environmental issue, necessitating effective sorting and categorization to facilitate efficient recycling practices. This study introduces a me...
详细信息
An Early Warning System (EWS) is an integrated system that disseminates early warning information to lessen a natural catastrophe's impact, facilitating preparedness and reaction processes. To reduce the likelihoo...
详细信息
Plant diseases pose severe risks to agricultural production and global food security. For prompt intervention and mitigation, early and precise disease detection is crucial. Due to the development of trustworthy compu...
详细信息
Based on a comparative analysis of the Long Short-Term Memory(LSTM)and Gated Recurrent Unit(GRU)networks,we optimize the structure of the GRU network and propose a new modulation recognition method based on feature ex...
详细信息
Based on a comparative analysis of the Long Short-Term Memory(LSTM)and Gated Recurrent Unit(GRU)networks,we optimize the structure of the GRU network and propose a new modulation recognition method based on feature extraction and a deep learning ***-order cumulant,Signal-to-Noise Ratio(SNR),instantaneous feature,and the cyclic spectrum of signals are extracted firstly,and then input into the Convolutional Neural Network(CNN)and the parallel network of GRU for *** modulation modes of communication signals are recognized *** results show that the proposed method can achieve high recognition rate at low SNR.
Early time classification algorithms aim to label a stream of features without processing the full input stream, while maintaining accuracy comparable to that achieved by applying the classifier to the entire input. I...
Early time classification algorithms aim to label a stream of features without processing the full input stream, while maintaining accuracy comparable to that achieved by applying the classifier to the entire input. In this paper, we introduce a statistical framework that can be applied to any sequential classifier, formulating a calibrated stopping rule. This data-driven rule attains finite-sample, distribution-free control of the accuracy gap between full and early-time classification. We start by presenting a novel method that builds on the Learn-then-Test calibration framework to control this gap marginally, on average over i.i.d. instances. As this algorithm tends to yield an excessively high accuracy gap for early halt times, our main contribution is the proposal of a framework that controls a stronger notion of error, where the accuracy gap is controlled conditionally on the accumulated halt times. Numerical experiments demonstrate the effectiveness, applicability, and usefulness of our method. We show that our proposed early stopping mechanism reduces up to 94% of timesteps used for classification while achieving rigorous accuracy gap control. Copyright 2024 by the author(s)
In actor-critic framework for fully decentralized multi-agent reinforcement learning (MARL), one of the key components is the MARL policy evaluation (PE) problem, where a set of N agents work cooperatively to evaluate...
详细信息
The lexicon is an essential component in the hybrid automatic speech recognition (ASR) system. However, a high-quality lexicon requires significant efforts from the linguistic experts and is difficult to obtain, espec...
详细信息
As the global population ages, neurocognitive disorders like the Alzheimer's disease which is a form of dementia, and several other forms of dementia are becoming more commonplace. These conditions significantly i...
详细信息
Cloud-based Deep learning of models has been widely applied in intelligent healthcare management to deliver improved diagnostics and analytics for advanced patient outcomes. The aim of this research is to assess the c...
详细信息
暂无评论