Research

Currently, I work on problems in causal inference, particularly on interference and experimentation in marketplaces. I also do research in neural networks and probability theory, particularly statistical mechanics.

Research

Optimizing the Returns to Experimentation Programs.
with S. Ejdemyr, A. Lal, M. Tingley.
Draft available upon request (2024). Accepted at CODE@MIT 2024

Regression Adjustments for Experimental Designs in Two-Sided Marketplaces.
with L. Lei, L. Masoero, S. Vijaykumar, G. Imbens, J. McQueen.
Work in progress (2024). Accepted at CODE@MIT 2024

Non-Identifiability distinguishes Neural Networks among Parametric Models.
S. Chatterjee*, T. Sudijono*.
Draft available upon request (2024).

Neural Networks Generalize on Low Complexity Data.
S. Chatterjee*, T. Sudijono*.
To be submitted (2024).
[ ArXiv ]

Synthetic Control Inference via Refined Placebo Tests.
L. Lei.*, T. Sudijono*,
Submitted (2024). Accepted at CODE@MIT 2023, ACIC 2024, CalMetrics 2024.
[ ArXiv | Software ]

Fluctuation Bounds in the Restricted Solid-on-Solid Model of Surface Growth.
T. Sudijono.
Submitted (2023).
[ ArXiv ]

A Topological Data Analytic Approach for Discovering Biophysical Signatures in Protein Dynamics.
W. S. Tang*, G. M. da Silva*, H. Kirveslahti, E. Skeens, B. Feng, T. Sudijono, K. Yang, S. Mukherjee, B. Rubinstein, L. Crawford.
PLOS Computational Biology (2022).
[ Journal | BioArXiv ]

A statistical pipeline for identifying physical features that differentiate classes of 3D shapes.
B. Wang*, T. Sudijono*, H. Kirveslahti*, T. Gao, D. M. Boyer, S. Mukherjee, and L. Crawford.
Annals of Applied Statistics (2021).
[ Journal | BioArXiv | Software ]

* denotes equal contribution or alphabetical order.

Theses

Stationarity and Ergodicity of Local Dynamics of Interacting Markov Chains on Large Sparse Graphs.
T. Sudijono, (Advised by K. Ramanan, A. Ganguly).
Sc. B. Thesis - May 2019.
[ Repository ]