The following is translated from the original:
As the basic technology of artificial intelligence, existing machine learning (ML) methods often rely on a large number of human interventions and manual presets, such as manually collecting, selecting and annotating data, manually building the infrastructure of deep neural networks, and determining the type of algorithm and its optimization. Hyperparameters of the algorithm, etc.
To solve these current challenges in machine learning, the research team at Xi’an Jiaotong University has developed a new method called Simulation Learning Method (SLeM). The core concept of SLeM is to simulate and extract ML learning methods traditionally set by humans and transform them into automated learning processes. Essentially, the SLeM framework represents an ML for ML paradigm in which ML tools are used to design and optimize the basic elements of ML.
The team developed a series of machine learning automation algorithms based on the SLeM framework, demonstrating their effectiveness in enhancing the adaptive learning capabilities of existing machine learning methods.
“Recently, many AutoML methods have been proposed to implement ML automation. However, most existing AutoML methods are heuristic in nature and it is difficult to establish a solid theoretical foundation. In contrast, the SLeM framework provides a unified mathematical formula for ML automation. Professor Xu Zongben, the first author of the paper and an academician of the China Academy of Sciences, said: “This provides theoretical insights for SLeM’s responsibility transfer generalization capabilities.”
The development of high-level large language models (LLMs) has become the cornerstone of artificial intelligence, significantly expanding the ability to solve a variety of applications and tasks. However, the ML community has not yet fully addressed the basic theoretical evidence for LLMs ‘superior responsibility generalization ability. The novel SLeM method provides a promising perspective and tool for advancing the research and understanding of responsibility generalization capabilities in large language models ( LLMs ).
If you want to learn more, you can click on the link below the video.
Thank you for watching this video. If you like it, please subscribe and like it. thank
Original text:https://techxplore.com/news/2024-08-simulating-methodology-approach-machine-automation.html
More information: Zongben Xu et al., Simulation Learning Method (SLeM): An automated method for machine learning, National Science Review (2024). DOI：10.1093/nsr/nwae277
Oil tubing: