Browse Articles

This study investigates how the hyperparameters epochs and batch size affect the classification accuracy of a convolutional neural network (CNN) trained on pulsar candidate data. Our results reveal that accuracy improves with increasing number of epochs and smaller batch sizes, suggesting that with optimized hyperparameters, high accuracy may be achievable with minimal training. These findings offer insights that could help create more efficient machine learning classification models for pulsar signal detection, with the potential of accelerating pulsar discovery and advancing astrophysical research.

Detecting cervical spondylosis using deep learning

The authors looked at the feasibility to predict wind speeds that will have less reliance on using historical data.

The authors looked at how soil temperature changes with fire to develop a sensor system that could aid in earlier detection of fires.

The authors looked at the ability of different machine learning algorithms to predict the level of financial corruption in different countries.

The authors looked at the ability of different deep learning models to predict the presence of lung cancer from chest CT scans. They found that a pre-trained CNN model performed better than an autoencoder model.

The authors looked at different ways to build computational resources that would analyze shooting form for basketball players.

Brain-Computer Interface (BCI) allows users, especially those with paralysis, to control devices through brain activity. This study explored using a custom transformer model to decode neural signals into handwritten text for individuals with limited motor skills, comparing its performance to a traditional RNN-based BCI.

The authors looked at different models of semantic segmentation to determine which may be best used in the future for segmentation of CT scans to help diagnose certain conditions.

Machine translation remains a challenging area in artificial intelligence, with neural machine translation (NMT) making significant strides over the past decade but still facing hurdles, particularly in translation quality due to the reliance on expensive bilingual training data. This study explores whether large language models (LLMs), like GPT-4, can be effectively adapted for translation tasks and outperform traditional NMT systems.