Machine Learning Research

Mohammadreza
Bakhtyari

Ph.D. Candidate in Computer Science

Université du Québec à Montréal (UQAM) · Montréal, Canada

Portrait of Mohammadreza Bakhtyari

About

I am a Ph.D. candidate in Computer Science at the Université du Québec à Montréal (UQAM), advised by Vladimir Makarenkov. I develop learning and inference algorithms for structured data, with a focus on probabilistic, permutation-aware neural models. I am interested in how synthetic priors, inductive biases, and learned representations enable transfer across datasets and tasks.

My doctoral research is supported by the Merit Scholarship Program for Foreign Students (PBEEE), awarded by FRQNT.

Current Work

My current research focuses on amortized inference for unsupervised learning and tabular foundation models. I study how models trained on synthetic tasks can generalize to new datasets, with an emphasis on clustering and probabilistic inference for heterogeneous tabular data.

Research interests

  • Amortized & probabilistic inference; prior-fitted networks (PFNs)
  • Meta-learning & in-context learning
  • Structured & permutation-aware neural architectures
  • Unsupervised learning

Selected Publications

Full list on Scholar

TabK: Amortized Bayesian Estimation of the Number of Clusters in Tabular Data

Mohammadreza Bakhtyari, Bogdan Mazoure, Renato Cordeiro de Amorim, Guillaume Rabusseau, Vladimir Makarenkov

40th Annual Conference on Neural Information Processing Systems (NeurIPS), 2026

Code

ClustRecNet: A Novel End-to-End Deep Learning Framework for Clustering Algorithm Recommendation

Mohammadreza Bakhtyari, Bogdan Mazoure, Renato Cordeiro de Amorim, Guillaume Rabusseau, Vladimir Makarenkov

IEEE Access, 2026

Paper Code

Soil Microbiome Prediction Using Traditional Machine Learning and Deep Learning Models

Zahia Aouabed, Vincent Therrien, Mohamed Achraf Bouaoune, Mohammadreza Bakhtyari, Mohamed Hijri, Vladimir Makarenkov

Scientific Reports, 2026

Paper

BayTTA: Uncertainty-Aware Medical Image Classification with Optimized Test-Time Augmentation Using Bayesian Model Averaging

Zeinab Sherkatghanad, Moloud Abdar, Mohammadreza Bakhtyari, Paweł Pławiak, Vladimir Makarenkov

Knowledge-Based Systems, 2025

Paper

EEG Motor Imagery Classification Based on a ConvLSTM Autoencoder Framework Augmented by Attention BiLSTM

Sayeh Mirzaei, Parisa Ghasemi, Mohammadreza Bakhtyari

Multimedia Tools and Applications, 2025

Paper

ADHD Detection Using Dynamic Connectivity Patterns of EEG Data and ConvLSTM with Attention Framework

Mohammadreza Bakhtyari, Sayeh Mirzaei

Biomedical Signal Processing and Control, 2022

Paper

Education

Ph.D. in Computer ScienceUniversité du Québec à Montréal (UQAM)
M.Sc. in Algorithms and ComputationsUniversity of Tehran