learning analytics has developed into a powerful tool for discovering unforeseen patterns in educational data and forecasting students' academic success. In order to anticipate the final exam grades of students st...
详细信息
Automation in manufacturing has played an important role in optimizing the welding process by attaining the utmost level of quality control with high operational efficiency. This paper analyzes 15,000 welding observat...
详细信息
Extensive use of computer-based automation in the healthcare sector has resulted in the collection of huge amounts of digital data. Due to which, healthcare professionals find it easy to examine symptoms precisely and...
详细信息
In companies or any organizations, the data is collected in various formats along with government-approved identity documents such as Aadhaar, PAN, license, etc. in soft or hard copy format. They contain valuable info...
详细信息
Grinding force is a key parameter in precision machining, significantly influencing material removal rates, surface quality, and machining vibrations. Traditional static force models often neglect the dynamic characte...
详细信息
Network traffic patterns in high-performance computing systems can impact application performance by inducing costly re-transmissions or consuming memory bandwidth. Emerging SmartNIC technologies present new opportuni...
详细信息
ISBN:
(纸本)9798350370621
Network traffic patterns in high-performance computing systems can impact application performance by inducing costly re-transmissions or consuming memory bandwidth. Emerging SmartNIC technologies present new opportunities for addressing these issues and optimizing network performance by providing a platform for the intelligent utilization of network resources through machinelearning models of network traffic. We analyze traffic data from scientific applications and proxies and propose lightweight approaches to modelling network traffic based on dynamic (i.e., rolling) linear regression and supplement it with random forest classifiers for additional accuracy, reaching 90% or higher.
Stance detection is a measure to understand the user view over the content posted on social media. Research associated with stance detection also includes modeling text to identify trends, opinions, feedback, reviews,...
详细信息
Global warming is a major problem today. Climate change is changing and human activity seems to be one of the main causes. A combination of open scientific data, freely provided by reputable organizations such as NASA...
详细信息
Due to the increasing degree of substation automation and intelligence and the promotion of unattended operation, the role of batteries is becoming more and more important. Battery life assessment methods usually rely...
详细信息
In this paper, we propose a novel approach called DIffusion-guided DIversity (DIDI) for offline behavioral generation. The goal of DIDI is to learn a diverse set of skills from a mixture of label-free offline data. We...
详细信息
In this paper, we propose a novel approach called DIffusion-guided DIversity (DIDI) for offline behavioral generation. The goal of DIDI is to learn a diverse set of skills from a mixture of label-free offline data. We achieve this by leveraging diffusion probabilistic models as priors to guide the learning process and regularize the policy. By optimizing a joint objective that incorporates diversity and diffusion-guided regularization, we encourage the emergence of diverse behaviors while maintaining the similarity to the offline data. Experimental results in four decision-making domains (Push, Kitchen, Humanoid, and D4RL tasks) show that DIDI is effective in discovering diverse and discriminative skills. We also introduce skill stitching and skill interpolation, which highlight the generalist nature of the learned skill space. Further, by incorporating an extrinsic reward function, DIDI enables reward-guided behavior generation, facilitating the learning of diverse and optimal behaviors from suboptimal data. Copyright 2024 by the author(s)
暂无评论