data mining and machinelearning techniques have been widely used in the knowledge extraction process of medical databases, one highlight being their use to improve diagnostic systems. Decision trees are supervised bl...
详细信息
Estimating influential parameters in real-world applications can be challenging, especially in problems where the model likelihood is intractable. To perform inference, the machinelearning model should be able to und...
详细信息
Heart disease is an illness that affects the heart and manifests itself in a variety of ways, including blood vessel issues, irregular heartbeats, and diseases of the heart muscle or valves. In recent years, researche...
详细信息
Wuhan, China was the place where COVID-19 was found in December 2019 and proved to be the worst pandemic in the world. This virus is highly contagious and spreads through direct contact and droplets. More than 210 cou...
详细信息
Vast volumes of data are readily available everywhere today. As a result, it is crucial to evaluate this data to draw out some relevant information and create algorithms based on that analysis. Algorithms are develope...
详细信息
In the present era of deep learning, continual learning research is mainly focused on mitigating forgetting when training a neural network with stochastic gradient descent on a non-stationary stream of data. On the ot...
详细信息
In the present era of deep learning, continual learning research is mainly focused on mitigating forgetting when training a neural network with stochastic gradient descent on a non-stationary stream of data. On the other hand, in the more classical literature of statistical machinelearning, many models have sequential Bayesian update rules that yield the same learning outcome as the batch training, i.e., they are completely immune to catastrophic forgetting. However, they are often overly simple to model complex real-world data. In this work, we adopt the meta-learning paradigm to combine the strong representational power of neural networks and simple statistical models' robustness to forgetting. In our novel meta-continual learning framework, continual learning takes place only in statistical models via ideal sequential Bayesian update rules, while neural networks are meta-learned to bridge the raw data and the statistical models. Since the neural networks remain fixed during continual learning, they are protected from catastrophic forgetting. This approach not only achieves significantly improved performance but also exhibits excellent scalability. Since our approach is domain-agnostic and model-agnostic, it can be applied to a wide range of problems and easily integrated with existing model architectures. Copyright 2024 by the author(s)
While there has been significant progress in the application of transfer learning within reinforcement learning, most existing research primarily focuses on online reinforcement learning, with few studies addressing t...
详细信息
Matrix factorization (MF), a popular unsupervised learning technique for data representation, has been widely applied in data mining and machinelearning. According to different application scenarios, one can impose d...
详细信息
This paper investigates the effect of midterm and short quizzes on the success scores at the end of the course through machinelearning methods. data from 3,427 students in the Faculty of Engineering and Natural Scien...
详细信息
Throughout the outbreak, COVID-19 has been a hot issue on Twitter, providing a forum for people to voice their ideas and perspectives. Unfortunately, there isn’t a thorough rundown of sentiment analysis in Twitter di...
详细信息
暂无评论