Deep Learning and LLM-based Methods Applied to Stellar Lightcurve Classification
Authors:
Yu-Yang Li,
Yu Bai,
Cunshi Wang,
Mengwei Qu,
Ziteng Lu,
Roberto Soria,
Jifeng Liu
Abstract:
Light curves serve as a valuable source of information on stellar formation and evolution. With the rapid advancement of machine learning techniques, it can be effectively processed to extract astronomical patterns and information. In this study, we present a comprehensive evaluation of deep-learning and large language model (LLM) based models for the automatic classification of variable star ligh…
▽ More
Light curves serve as a valuable source of information on stellar formation and evolution. With the rapid advancement of machine learning techniques, it can be effectively processed to extract astronomical patterns and information. In this study, we present a comprehensive evaluation of deep-learning and large language model (LLM) based models for the automatic classification of variable star light curves, based on large datasets from the Kepler and K2 missions. Special emphasis is placed on Cepheids, RR Lyrae, and eclipsing binaries, examining the influence of observational cadence and phase distribution on classification precision. Employing AutoDL optimization, we achieve striking performance with the 1D-Convolution+BiLSTM architecture and the Swin Transformer, hitting accuracies of 94\% and 99\% correspondingly, with the latter demonstrating a notable 83\% accuracy in discerning the elusive Type II Cepheids-comprising merely 0.02\% of the total dataset.We unveil StarWhisper LightCurve (LC), an innovative Series comprising three LLM-based models: LLM, multimodal large language model (MLLM), and Large Audio Language Model (LALM). Each model is fine-tuned with strategic prompt engineering and customized training methods to explore the emergent abilities of these models for astronomical data. Remarkably, StarWhisper LC Series exhibit high accuracies around 90\%, significantly reducing the need for explicit feature engineering, thereby paving the way for streamlined parallel data processing and the progression of multifaceted multimodal models in astronomical applications. The study furnishes two detailed catalogs illustrating the impacts of phase and sampling intervals on deep learning classification accuracy, showing that a substantial decrease of up to 14\% in observation duration and 21\% in sampling points can be realized without compromising accuracy by more than 10\%.
△ Less
Submitted 16 April, 2024;
originally announced April 2024.
A Lossless Data Hiding Technique based on AES-DWT
Authors:
Francisco Rubén Castillo Soria,
Gustavo Fernández Torres,
Ignacio Algredo-Badillo
Abstract:
In this paper we propose a new data hiding technique. The new technique uses steganography and cryptography on images with a size of 256x256 pixels and an 8-bit grayscale format. There are design restrictions such as a fixed-size cover image, and reconstruction without error of the hidden image. The steganography technique uses a Haar-DWT (Discrete Wavelet Transform) with hard thresholding and LSB…
▽ More
In this paper we propose a new data hiding technique. The new technique uses steganography and cryptography on images with a size of 256x256 pixels and an 8-bit grayscale format. There are design restrictions such as a fixed-size cover image, and reconstruction without error of the hidden image. The steganography technique uses a Haar-DWT (Discrete Wavelet Transform) with hard thresholding and LSB (Less Significant Bit) technique on the cover image. The algorithms used for compressing and ciphering the secret image are lossless JPG and AES, respectively. The proposed technique is used to generate a stego image which provides a double type of security that is robust against attacks. Results are reported for different thresholds levels in terms of PSNR.
△ Less
Submitted 11 December, 2012;
originally announced December 2012.
