News

XGBoost is a widely used gradient-boosting framework in machine learning. Due to its low resource consumption and interpretability for sequential learning, XGBoost has gained increasing attention in green learning, where computational efficiency and model transparency are essential. However, XGBoost is not optimized in the current green learning literature. It accounts for a large part of the model parameters in a green learning system. It is desired to reduce the size of XGBoost while maintaining its high performance.

We recently have proposed a new method called Multi-stage XGBoost (MXB), a modular framework that builds a sequence of XGBoost models. MXB adopts a sequence of shallow XGBoost models, each of which defines a stage. Unlike conventional XGBoost, which is trained on the full feature set, the model at each stage of MXB is trained on a distinct feature subset. The general idea is to put more discriminant ones in earlier stages and less discriminant ones in later stages. This is motivated by the convergence curve of the XGBoost classifier and regressor. Also, the transition from one XGBoost model to the next in MXB is governed by the behavior of the gradient and Hessian, which represent the first- and second-order derivatives of the objective function, respectively. The gradient quantifies the direction and rate of change of the loss function, indicating how far the model is from optimality. The Hessian, on the other hand, measures the curvature of the loss, offering insight into the local stability of convergence.

We evaluate MXB on two datasets: MNIST and Fashion-MNIST, and compare it with standard XGBoost models of equivalent depth and model size. The training and testing curves show that MXB has smaller train–validation gaps and more stable testing loss as compared to [...]

We are very happy to welcome a new MCL member, Alek Yegazarian. Here is a quick interview with Alek:

1.  Could you briefly introduce yourself and your research interests?

My name is Alek Yegazarian, I am a recent Master of Science Electrical Engineering graduate from the University of Southern California, earning the Ming Hsieh Department of Electrical and Computer Engineering academic achievement award and the 2025 Viterbi Master’s 1 student commencement speaker honor. My research background and interests lie at the intersection of computational imaging, computer vision, and robotics. Outside of engineering and academia, I am a magician member of the Academy of Magical Arts at the Magic Castle in Hollywood, enjoy playing piano, and stay active by practicing taekwondo.2. What is your impression of MCL and USC?

I am so excited and grateful to be a part of MCL. Dr. Kuo’s passion and enthusiasm for his students and research is inspiring. I especially enjoy the lab environment, as everyone is very friendly and supportive. I am truly happy and excited every time I come to campus for research! Shout out to Wei Wang and Yixing Wu for being so welcoming, attentive, and helpful during these first few weeks working with them.

3. What is your future expectation and plan in MCL?

During this summer, I am supporting research on Green U-Shaped Learning (GUSL) for super-resolution and denoising tasks. I hope to learn and contribute as much as I can to MCL and Green Learning, while simultaneously getting to know my fellow lab members and getting a taste of life as a researcher in academia.

We are very happy to welcome a new MCL member, Kevin Lim. Here is a quick interview with Kevin:

1. Could you briefly introduce yourself and your research interests?

My name is Kevin Lim, and I am a second year master’s student studying Computer Science. I’ve explored a few subfields within the field of AI but I’m most interested in deep learning and computer vision. I found Professor Kuo’s work in green learning to be very interesting which is part of my motivation to apply for an internship at MCL.

2. What is your impression of MCL and USC?

I think USC is a place with many opportunities for ambitious people. Hard-working and passionate individuals thrive here and this serves as an inspiration for others as well. I’ve spoken to a few people at MCL already and they are very friendly, and I appreciate Professor Kuo’s leadership and guidance.

3. What is your future expectation and plan in MCL?

I hope to gain a better understanding of research trends in the image processing field, as I’ve already started to do while preparing for the Mediatek research project. I also hope to build strong connections with members of the lab.

Image classification is a key task in computer vision, typically driven by high-performing deep learning methods. However, these methods are criticized for lacking transparency. As an alternative, Green Learning offers more interpretable models, though their performance is not as strong as that of deep learning.

We propose a novel Green Learning framework that enhances performance by integrating label supervision directly into the feature extraction process. A key component of our approach is the LDA filter block, which is a feedforward mechanism that uses Linear Discriminant Analysis (LDA) to create convolution filters without the need for backpropagation. Additionally, we present spectral LNT, a new variant of the least-squares normal transform (LNT) that takes advantage of the spatial-spectral structure of feature maps by applying localized linear combinations of features. Together, our methods create a label-guided, interpretable feature extraction pipeline.

Congratulations to Qingyang Zhou on successfully defending his dissertation today! His thesis, titled “Advanced Techniques for Point Cloud Quality Assessment, Surface Reconstruction, and Coding,” was reviewed by his committee—chaired by Jay Kuo, with Antonio Ortega as a member and Stefanos Nikolaidis serving as the outside examiner. The committee praised the rigor and excellence of Qingyang’s research. Here is a summary of his thesis:Point clouds (PCs) play a critical role in 3D vision and graphics applications such as AR/VR, digital preservation, and medical imaging. However, their large data volume, vulnerability to quality degradation, and the need for surface reconstruction from sparse, noisy points pose long-standing challenges. In this work, we address these issues through three research thrusts: compression, quality assessment, and surface reconstruction. First, we propose GPCGC, a low-complexity geometry compression method that leverages vector quantization and block-level rate-distortion modeling to achieve fast encoding/decoding with competitive performance. Second, we introduce GPQA, an interpretable, saliency-guided quality assessment framework that projects local 3D structures into 2D patches and uses a green machine learning model to predict perceptual quality under both full- and no-reference settings. Third, we develop two surface reconstruction methods: GPSR, an unsupervised diffusion-based approach, and LPSR, a supervised variant under the green learning paradigm. Together, these contributions advance the efficiency, interpretability, and robustness of the point cloud processing pipeline, enabling practical deployment in resource-constrained environments.He generously shared his experiences in the MCL Lab with us:My PhD journey at the University of Southern California’s Media Communications Lab (MCL) has been one of the most rewarding chapters of my life. Under the guidance of Professor C.-C. Jay Kuo, and with the generous support of MCL alumni and lab members, I not only deepened my technical knowledge but [...]