Nima Pourdamghani

Nima Pourdamghani

PhD Candidate, University of Southern California

My Resume

About Me

I am a PhD candidate in the computer science department at University of Southern California (USC). I am working with Professor Kevin Knight at the Information Sciences Institute (ISI). Prior to joining ISI, I got my Bachelor and Masters degree from Sharif University of Technology.

My research interests are in the fields of Natural Language Processing, Machine Learning and Artificial Intelligence. Recently, I have been working on machine translation for low resource languages where we extend automatic translation technology to hundreds of languages that are not covered by the current machine translation systems.


Natural Language Processing

  • "Neighbors Helping the Poor: Improving Low Resource Machine Translation Using Related Languages"
    (N. Pourdamghani, K. Knight), submitted to Machine Translation, 2018

  • "Using Word Vectors to Improve Word Alignments for Low Resource Machine Translation"
    (N. Pourdamghani, M. Ghazvininejad, K. Knight), Proc. NAACL, 2018

  • "Incident-Driven Machine Translation and Name Tagging for Low-resource Languages"
    (U. Hermjakob, [et al, including N. Pourdamghani]), Machine Translation, 2018 [pdf]
  • "Deciphering Related Languages"
    (N. Pourdamghani, K. Knight), Proc. EMNLP, 2017. [pdf]

  • "Team ELISA System for DARPA LORELEI Speech Evaluation 2016"
    (P. Papadopoulos, [et al, including N. Pourdamghani]), Proc. Interspeech, 2017. [pdf]
  • "Generating English from Abstract Meaning Representations"
    (N. Pourdamghani, K. Knight), Proc. INLG, 2016. [pdf] [code]

  • "A Multi-media Approach to Cross-lingual Entity Knowledge Transfer"
    (D. Lu, X. Pan, N. Pourdamghani, H. Ji, SF. Chang, K. Knight), Proc. ACL, 2016. [pdf]
  • "Aligning English Strings with Abstract Meaning Representation Graphs",
    (N. Pourdamghani, Y. Gao, U. Hermjakob, K. Knight), Proc. EMNLP, 2014. [pdf] [gold alignments] [code]

Semi-Supervised Learning + Computer Vision

  • "Graph based semi-supervised human pose estimation: When the output space comes to help"
    (N. Pourdamghani, H. R. Rabiee, F. Faghri, M. H. Rohban), Pattern Recognition Letters, 2012. [pdf]

  • "Metric learning for graph based semi-supervised human pose estimation"
    (N. Pourdamghani, H. R. Rabiee, M. Zolfaghari), Proc. ICPR, 2012. [pdf]

  • "Multi-directional spatial error concealment using adaptive edge thresholding"
    (H. Asheri, H. R. Rabiee, N. Pourdamghani, M. Ghanbari), IEEE Trans. on Consumer Electronics, 2012. [pdf]

  • "HMM based semi-supervised learning for activity recognition"
    (M. Ghazvininejad, H. R. Rabiee, N. Pourdamghani, P. Khanipour), Proc. SAGAWARE, Ubicomp, 2011. [pdf]

  • "A gaussian process regression framework for spatial error concealment with adaptive kernels"
    (H. Asheri, H. R. Rabiee, N. Pourdamghani, M. H. Rohban), Proc. ICPR, 2010. [pdf]


project name

AMR: Abstract Meaning Representation

The AMR Bank is a set of English sentences paired with simple, readable semantic representations. We hope that it will spur new research in natural language understanding, generation, and translation. Please visit the AMR page for details.

project name

3D Human Pose Estimation

In this project we propose methods to infer a full-body 3D pose of humans from a single image captured by a typical camera. Human pose estimation plays a key role in many applications such as video surveillance systems, human computer interaction, interactive games, and special effects industry.

project name

Video Error Concealment

Transmission of image and video signals over error-prone channels is susceptible to bit erasures and packet losses. In this project we use correctly received areas of the images to recover those areas corrupted by transmission errors.