Mohammadreza Armandpour

Statistics Ph.D. Student at Texas A&M University
armand@stat.tamu.edu
GitHub, google scholar


I am a Statistics Ph.D. student at Texas A&M University, advised by Dr. Mingyan Zhou at the University of Texas at Austin. I am also honored to be co-advised by Debdeep Pati and Jianhua Huang at the Texas A&M University.

I am mainly interested in developing new approaches to generative modeling that allows stable training algorithms and high-quality sample generation. I also have worked on advancing ML methods that can efficiently and effectively handle graph-structured data. During my Ph.D., I had amazing collaborators, including Chunyuan Li (Microsoft Research), Ashish Shrivastave (Apple MIND team), Tracy Ke (Harvard University), Ali Sadeghian (University of Florida), and Xia Hu (Texas A&M University).

Previously, I did my undergraduate studies in Electrical Engineering and Mathematics (double majors) at the Sharif University of Technology, where I also did my Masters in Applied Mathematics. Outside of research, I enjoy hiking and reading biographies and history.

I interned at Apple (MIND team) as a machine learnig researcher. I also have done internships with Tesla (Autopilot team) (Spring 2021) and Eli Lilly and Company (Summer 2019). During my Ph.D. second year, I had the opportunity to work at the DATA (Data Analytics at Texas A&M) Lab, supervised by Xia Hu .

My thesis title is "Deep Generative Models: Pitfalls and Fixes", and it can be found Here.

Curriculum Vitae (CV)

Publications


Bayesian Graph Contrastive Learning
Arman Hasanzadeh, Mohammadreza Armandpour, Ehsan Hajiramezanali, Mingyuan Zhou, Nick Duffield, Krishna Narayanan
ICML 2022 (under review)

Synt++: Utilizing Imperfect Synthetic Data To Improve Speech Recognition
Ting-Yao Hu, Mohammadreza Armandpour, Ashish Shrivastava, Jen-Hao Rick Chang, Hema Koppula, Oncel Tuzel
ICASSP 2021

Deep Spatio-Temporal Wind Power Forecasting
Jianyuan Li, Mohammadreza Armandpour
ICASSP 2021 code

Convex Polytope Trees and Its Application to VAE
Mohammadreza Armandpour, Ali Sadeghian, Mingyaun Zhou
NeurIPS 2021, implemented in PyTorch, code

Partition-Guided GANs
Mohammadreza Armandpour, Ali Sadeghian, Chunyuan Li, Mingyaun Zhou
CVPR 2021, implemented in PyTorch, code .

ChronoR: Rotation Based Temporal Knowledge Graph Embedding
Ali Sadeghian*, Mohammadreza Armandpour (co-first author)*, Anthony Collas, Daisy Zhe Wang
AAAI 2021 , implemented in PyTorch.

Convex Polytope Trees
Mohammadreza Armandpour, Mingyaun Zhou
Under Review, implemented in PyTorch, code .

Deep Personalized Glucose Level Forecasting Using Attention-based Recurrent Neural Networks
Mohammadreza Armandpour, Brian Kidd, Yu Du, Jianhua Huang
IJCNN 2021, implemented in PyTorch, the code will be released soon.

DRUM: End-To-End Differentiable Rule Mining On Knowledge Graphs
Ali Sadeghian*, Mohammadreza Armandpour (co-first author)*, Patrick Ding, Daisy Zhe Wang
Advances in Neural Information Processing Systems (NeurIPS 2019) acceptance rate: 21.1% (1,428 accepted out of 6,743 submissions), implemented in TensorFlow, code .

Robust Negative Sampling for Network Embedding
Mohammadreza Armandpour, Patrick Ding, Jianhua Huang, Xia Hu
Association for the Advancement of Artificial Intelligence (AAAI 2019), acceptance rate 16.2% (1,150/ 7,095 ).

Awards


Gold Medal in Iranian National Mathematical Olympiad
The medal is awarded annually to 12 individuals out of 320,000 students, 2007