Rasul Kairgeldin

PhD Candidate Electrical Engineering and Computer Science (EECS) School of Engineering University of California, Merced 5200 N. Lake Road Merced, CA 95343 Email: rkairgeldin [at] ucmerced.edu WWW: http://rkairgeldin.github.io LinkedIn Google Scholar

Research interests

My research focus is on Machine Learning and Optimization, specifically tree-based models and neural network compression. My recent projects are on the topic of tree-based models in various domains: supervised learning, clustering, neural network compression, vector quantization and model interpretability.

About

I am a 4th year PhD candidate at University of California, Merced (UCM) supervised by Miguel Á. Carreira-Perpiñán. I hold a MSc in Computer Engineering and BEng in Electrical and Electronic Engineering from NU in Astana, Kazakhstan. You can find my CV [here] .

Publications

[Google Scholar page]

[ICIP] R. Kairgeldin and M. Á. Carreira-Perpiñán (2025): "Fast image vector quantization using sparse oblique regression trees".
IEEE International Conference on Image Processing (ICIP 2025), Sep. 14 - 18, 2025.
[external link] [paper preprint] [slides] [video] [poster]
[NeSy] R. Kairgeldin and M. Á. Carreira-Perpiñán (2025): "Neurosymbolic models based on hybrids of convolutional neural networks and decision trees".
19th International Conference on Neurosymbolic Learning and Reasoning (NeSy 2025), Sep. 8 - 10, 2025.
[external link] [external link2] [paper preprint] [slides] [poster]
[KDD] R. Kairgeldin and M. Á. Carreira-Perpiñán (2024): "Bivariate decision trees: smaller, interpretable, more accurate".
Proceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD 2024), Aug. 25 - 29, 2024.
[external link] [paper preprint] [slides] [video] [poster]

[Workshop at NeurIPS] R. Kairgeldin and M. Á. Carreira-Perpiñán (2024): "Bivariate decision trees: smaller, interpretable, more accurate".
Workshop on Interpretable AI: Past, Present and Future (NeurIPS 2024), Dec. 15, 2024.
[external link] [paper preprint] [poster]
[UAI] R. Kairgeldin and M. Gabidolla and M. Á. Carreira-Perpiñán (2024): "Adaptive Softmax Trees for Many-Class Classification".
40th Conference on Uncertainty in Artificial Intelligence (UAI 2024), July 15 - 19, 2024.
[external link] [external link2] [paper preprint] [poster]
[BayLearn] R. Kairgeldin and M. Gabidolla and M. Á. Carreira-Perpiñán (2024): "Adaptive Softmax Trees for Many-Class Classification".
Bay Area Machine Learning Symposium (BayLearn 2024), Oct. 10, 2024
[external link] [paper preprint] [poster]
[BayLearn] R. Kairgeldin and M. Á. Carreira-Perpiñán (2023): "Bivariate decision trees".
Bay Area Machine Learning Symposium (BayLearn 2023), Oct. 19, 2023
[external link] [paper preprint] [poster]

Industry Experience

PhD ML Engineering Intern (Ranking & Foundational AI) at Meta. Worked on user and ad embedding learning for large Deep Recommender Systems. Mentors: Dave Li.
PhD ML Engineering Intern (Core ML) at Meta. Worked on compressing and accelerating Deep Recommender Systems on next-generation Meta Training and Inference Accelerator (MTIA). Mentors: Deboleena Roy.

Professional Activities

Reviewer, Neural Information Processing Systems (NeurIPS 2024, 2025).
Reviewer, International Conference on Machine Learning (ICML 2025).
Reviewer, International Conference on Learning Representations (ICLR 2025).
Reviewer, Journal on Machine Learning Research Conference on Learning Representations (JMLR 2024).
Reviewer, ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD 2024).
Reviewer, The international journal Advances in Data Analysis and Classification (ADAC 2022).
Student Volunteer, 30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD 2024).

Awards

KDD 2024 Student Travel Award ($1000).
NeSy 2025 Travel Grant Award ($500).