Amir Ghasemian

AmirGh2.jpg

Computational Social Science Lab

University of Pennsylvania

Philadelphia, PA 19104
Email me

I am a research scientist in Computational Social Science Lab at the University of Pennsylvania, working with Duncan Watts. Previously, I was a CIFellow 2020 in HNL at Yale University, working with Nicholas Christakis and Edoardo Airoldi. I received my PhD degree in Computer Science from University of Colorado Boulder under the supervision of Aaron Clauset.

My research interests lie in network science, statistical inference, causal inference, information theory, machine learning, data mining, and signal processing.

My Research


My current research involves inference and learning in complex networks. More specifically I am interested in looking at inference problems from different angles like Bayesian approaches in statistics, free energy in statistical physics and information theoretic perspective from electrical engineering and physics. I am interested in complex networks analysis and modeling using probabilistic graphical models, machine learning techniques, non-parametric models, information theory, signal processing, optimization, and linear algebra. I like to use different perspectives to analyze the same problem to understand better the problem and find the relations among different perspectives. This is the reason that I am interested to interdisciplinary areas in computer science. Mostly I would like to design and analyze models for these problems and investigate under what conditions we can have informative solutions. I am interested to statistical physics because of its unique perspective of phase transitions in making conjectures, to probability theory to find proofs of these conjectures, to statistics and information theory to model and analyze problems, and to linear algebra and signal processing to analyze and understand the geometry of the data available for problems.


news

May 15, 2024 My paper, “The Structure and Function of Antagonistic Ties in Village Social Networks,” has been accepted for publication in PNAS.
Jan 18, 2024 Our paper, “Sequential Stacking Link Prediction Algorithms for Temporal Networks,” has been accepted for publication in Nature Communications.
Jan 1, 2024 I joined Computational Social Science Lab at the University of Pennsylvania as a research scientist.
Dec 14, 2023 Our paper, “Causally estimating the effect of YouTube’s recommender system using counterfactual bots,” has been accepted for publication in PNAS.
Nov 9, 2023 My paper, “The Enmity Paradox,” has been accepted for publication in Scientific Reports.

selected publications

  1. PNAS-04c.png
    The structure and function of antagonistic ties in village social networks
    Amir Ghasemian, and Nicholas A Christakis
    Proceedings of the National Academy of Sciences, 2024
    Image courtesy of Cavan Huang
  2. DALLE2024-07-2515.18.50.png
    The Enmity Paradox
    Amir Ghasemian, and Nicholas A Christakis
    Scientific Reports, 2023
  3. model_stacking.png
    Stacking models for nearly optimal link prediction in complex networks
    Amir Ghasemian, Homa Hosseinmardi, Aram Galstyan, and 2 more authors
    Proceedings of the National Academy of Sciences, 2020
  4. over_under.jpg
    Evaluating overfit and underfit in models of network community structure
    Amir Ghasemian, Homa Hosseinmardi, and Aaron Clauset
    IEEE Transactions on Knowledge and Data Engineering, 2019
  5. PhysRevX.6.031005.png
    Detectability thresholds and optimal algorithms for community structure in dynamic networks
    Amir Ghasemian, Pan Zhang, Aaron Clauset, and 2 more authors
    Physical Review X, 2016