Amir-768x838.jpg

I am an AI scientist (Ph.D. from Carnegie Mellon University) and a computer engineer (B.S. from University of Tehran). My research is focused on pushing the frontiers of multimodality in machine learning. I am also interested in fundamental unimodal research topics including facial and body analysis, audio source localization and separation, as well as language and dependency modeling. During my Ph.D. I was a recipient of Yahoo Fellowship, and two honorable mention paper awards at ICMI. I was also an invitee to the White House Frontiers Conference. I have published, reviewed, and organized workshops at top ML conference.

resume.jpg
images.png
kisspng-google-scholar-education-researc

Selected Publications

The following papers are representitive of my research direction:

​

  • [2021] Relay Variational Inference: A Method for Accelerated Encoderless VI (preprint)

  • [2020] Variational Auto-Decoder: Neural Generative Modeling from Missing Data (preprint)

  • [2019] Social-IQ: A Question Answering Benchmark for Artificial Social Intelligence (CVPR, Oral)

  • [2018] Multimodal Language Analysis: CMU-MOSEI and Interpretable Dynamic Fusion Graph (ACL, Oral)

  • [2017] Tensor Fusion Network for Multimodal Sentiment Analysis (EMNLP, Oral)

​

News

  • [2021] Ph.D. Thesis: Robust Multimodal Learning from Language, Vision and Acoustic Modalities.

Special thanks to my committee: LP Morency, Ruslan Salakhutdinov, Tom Mitchel and Ali Farhadi.

  • [2021] NAACL: I organized the Third Workshop on Multimodal AI.

  • [2021] CVPR: I gave a tutorial on Multimodal Language.

  • [2021] NAACL: One paper accepted.

  • [2021] Preprint: StarNet: Gradient-free Training of Deep Generative Models using Determined System of Linear Equations.

  • [2021] AAAI: One paper accepted.

  • [2021] ICMI: Two papers accepted.

  • [2021] Information Fusion (IF 12.97): One paper accepted.

​

​

  • [2020] EMNLP: One paper accepted - CMU-MOSEAS dataset released.

  • [2020] Information Fusion (IF 12.97): Foundations of Multimodal Co-learning.

  • [2020] ACL: One paper accepted - Multimodal Finetuning of Advanced Transformer Models.

  • [2020] Preprint: Pseudo-encoded Stochastic Variational Inference.

  • [2020] Preprint: Variational Auto-Decoder: Neural Generative Modeling from Partial data. 

  • [2020] CVPR : Organizer of Multimodal Learning (MuL) Workshop.

  • [2020] ACL: Organizer of Second Grand-challenge and Workshop on Multimodal Language (proceedings). 

​

​

  • [2019] ICCV: Organizer of Cross-Modal Learning in the Wild (CroMol)

  • [2019] CVPR: One paper accepted as oral presentation.

  • [2019] ICLR: One paper accepted.

  • [2019] Invited Talk: Google San Francisco Workshop on Multimodal Learning.

  • [2019] AAAI: Session Chair of Deep Learning session, and NLP session.

  • [2019] AAAI: One paper accepted.

​

​

  • [2018] EMNLP: One paper accepted as oral presentation.

  • [2018] ACL: Two papers accepted as oral presentation.

  • [2018] AAAI: Two papers accepted as oral presentation.

  • [2018] ACL: Organizer of First Grand-challenge and Workshop on Multimodal Language (proceedings)

​

​

  • [2017] EMNLP: One paper accepted as oral presentation.

  • [2017] ACL: Two papers accepted (one oral).

​

​

  • [2016] Yahoo InMind Fellowship.

  • [2016] Invited to the White House Frontiers Conference.