I am an Applied Scientist at VCV-Science group at Apple where I work on vision foundation models. Prior to that, I was a CS Ph.D. student and graduate researcher at Oregon State University supervised by Prof. Fuxin Li.
My research interests mainly include foundation models, generative models, 2D/3D computer vision, interpretability, long-tail problem, self-supervised learning.
My research interests mainly include foundation models, generative models, 2D/3D computer vision, interpretability, long-tail problem, self-supervised learning.
News [outdated]
- Mar 22: Our counterfactual visual explanation paper "Cycle-Consistent Counterfactuals by Latent Transformations" (C3LT), is accepted to the CPVR 2022! (acceptance rate 25.33%) [PDF] [Github]
- Oct 21: Our paper "From Heatmaps to Structural Explanations of Image Classifiers" is accpeted to the Applied AI Journals! [PDF]
- May 21: Our paper "Re-Understanding Finite-State Representations of Recurrent Policy Networks" is accpeted to the ICML 21! (acceptance rate 21.4%) [PDF]
- May 21: Our new paper on flexible training of deep networks using ADMM titled "Stochastic Block-ADMM for Training Deep Networks " is now available on arXiv. [PDF]
- Feb 21: Code for iGOS++ is released! This repo also includes faster version of I-GOS! [Github]
- Feb 21: Our paper "iGOS++: integrated gradient optimized saliency by bilateral perturbations" is accepted to the ACM-CHIL 21! iGOS++ shows significant imporvement over I-GOS, particularly in terms of insertion score! (acceptnace rate ~27%) [PDF]
- Jan 21: I joined Scale AI as a Machine Learning Research Engineer Intern starting January 2021.
- Nov 20: Our paper "Embedding Deep Networks into Visual Explanations" is accepted to the Artificial Intelligence Journal! [Journal]
- Jul 20: Full-length paper for "Understanding Finite-State Representations of Recurrent Policy Networks" is now available at arXiv! [PDF]
- Jul 20: Our paper "Understanding Finite-State Representations of Recurrent Policy Networks" is accpeted to the ICML 20 XXAI workshop!
- Jun 20: My M.Sc. dissertation "Toward Disentangling the Activations of the Deep Networks via Low-dimensional Embedding and Non-negative Factorization" is now publicly available! [PDF]
- Apr 20: Demo for I-GOS is updated! It now supports user-input images, comparison against GradCam, and interactive deletion/insertion games! Go check it out! [Demo v2.0]
- Mar 20: I successfully passed my Ph.D. qualifying exam and defended my M.Sc. in Computer Science at Oregon State University! Yay! [Event]
- Feb 20: Medium post for I-GOS is now available! [Medium]
- Nov 19: Our paper "Visualizing Deep Networks by Optimizing with Integrated Gradients" is accepted to the AAAI 20. (acceptance rate 20.6%) [PDF]
- Aug 19: Code for I-GOS is now available! It is easy to use, go check it out! [Github]
- Jul 19: Early-version demo for I-GOS is no available! It does not yet support user-input images due to GPU constraints. Tune in for updtaes when more resources is available! [Demo]
- Jun 19: I-GOS is accepted to the CVPR 19 Explainable AI workshop. [PDF]
Publications
-
Mehran Soltani, Mohammad Hasan Shammakhi, Saeed Khorram, and Hamid Sheikhzadeh. ”Combined mRMR filter and sparse Bayesian classifier for analysis of gene expression data.”, International Conference of Signal Processing and Intelligent Systems (ICSPIS), IEEE, 2016. [PDF]
-
Mohamadreza Jafaryani, Saeed Khorram, Vahid Pourahmadi, and Minoo Shahbazi. “Sleep Stage Scoring Using Joint Frequency-Temporal and Unsupervised Features,”, International Conference on New Research Achievements in Electrical and Computer Engineering (ICNRAECE), IEEE, 2016. [PDF]