Redirigiendo al acceso original de articulo en 24 segundos...
Inicio  /  Algorithms  /  Vol: 15 Par: 9 (2022)  /  Artículo
ARTÍCULO
TITULO

High Per Parameter: A Large-Scale Study of Hyperparameter Tuning for Machine Learning Algorithms

Moshe Sipper    

Resumen

Hyperparameters in machine learning (ML) have received a fair amount of attention, and hyperparameter tuning has come to be regarded as an important step in the ML pipeline. However, just how useful is said tuning? While smaller-scale experiments have been previously conducted, herein we carry out a large-scale investigation, specifically one involving 26 ML algorithms, 250 datasets (regression and both binary and multinomial classification), 6 score metrics, and 28,857,600 algorithm runs. Analyzing the results we conclude that for many ML algorithms, we should not expect considerable gains from hyperparameter tuning on average; however, there may be some datasets for which default hyperparameters perform poorly, especially for some algorithms. By defining a single hp_score value, which combines an algorithm?s accumulated statistics, we are able to rank the 26 ML algorithms from those expected to gain the most from hyperparameter tuning to those expected to gain the least. We believe such a study shall serve ML practitioners at large.

Palabras claves

 Artículos similares

       
 
Håkan Emteborg and Jean Charoud-Got    
A high-resolution infrared (IR) camera was used for temperature measurements in a pharmaceutical formulation (mannitol/sucrose solution, 4:1%, m/m) during a freeze-drying process. The temperature was measured simultaneously at the surface as well as vert... ver más
Revista: Applied Sciences

 
Naseem Adnan Alsamarai and Osman Nuri Uçan    
Today, the IoT has become a vital part of our lives because it has entered into the precise details of human life, like smart homes, healthcare, eldercare, vehicles, augmented reality, and industrial robotics. Cloud computing and fog computing give us se... ver más
Revista: Applied Sciences

 
Leonard Vance, Agustin Espinoza, Jorge Martinez Dominguez, Salil Rabade, Gavin Liu and Jekan Thangavelautham    
Sustainable space exploration will require using off-world resources for propellant generation. Using off-world-generated propellants significantly increases future missions? range and payload capacity. Near Earth Objects (NEOs) contain a range of availa... ver más
Revista: Aerospace

 
Tianhao Wang, Hongying Meng, Rui Qin, Fan Zhang and Asoke Kumar Nandi    
Wind turbines are a crucial part of renewable energy generation, and their reliable and efficient operation is paramount in ensuring clean energy availability. However, the bearings in wind turbines are subjected to high stress and loads, resulting in fa... ver más
Revista: Applied Sciences

 
Guohui Xu, Shiqing Sun, Yupeng Ren, Meng Li and Zhiyuan Chen    
Turbidity currents are important carriers for transporting terrestrial sediment into the deep sea, facilitating the transfer of matter and energy between land and the deep sea. Previous studies have suggested that turbidity currents can exhibit high velo... ver más