CausalSim: A Causal Framework for Unbiased
Trace-Driven Simulation

Abdullah Alomar2,3*      Pouya Hamadanian1*      Arash Nasr-Esfahany1*      Anish Agarwal2,3     
Mohammad Alizadeh1      Devavrat Shah2,3     

1Computer Science and Artificial Intelligence Laboratory (MIT CSAIL)
2Institute for Data, Systems and Society (MIT IDSS)
3Laboratory for Information and Decision Systems (MIT LIDS)


We present CausalSim, a causal framework for unbiased trace-driven simulation. Current trace-driven simulators assume that the interventions being simulated (e.g., a new algorithm) would not affect the validity of the traces. However, real-world traces are often biased by the choices algorithms make during trace collection, and hence replaying traces under an intervention may lead to incorrect results. CausalSim addresses this challenge by learning a causal model of the system dynamics and latent factors capturing the underlying system conditions during trace collection. It learns these models using an initial randomized control trial (RCT) under a fixed set of algorithms, and then applies them to remove biases from trace data when simulating new algorithms.

Key to CausalSim is mapping unbiased trace-driven simulation to a tensor completion problem with extremely sparse observations. By exploiting a basic distributional invariance property present in RCT data, CausalSim enables a novel tensor completion method despite the sparsity of observations. Our extensive evaluation of CausalSim on both real and synthetic datasets, including more than ten months of real data from the Puffer video streaming system shows it improves simulation accuracy, reducing errors by 53% and 61% on average compared to expert-designed and supervised learning baselines. Moreover, CausalSim provides markedly different insights about ABR algorithms compared to the biased baseline simulator, which we validate with a real deployment.



CausalSim: A Causal Framework for Unbiased Trace-Driven Simulation
Abdullah Alomar*, Pouya Hamadanian*, Arash Nasr-Esfahany*, Anish Agarwal, Mohammad Alizadeh, Devavrat Shah
Proceedings of the 20th USENIX Symposium on Networked Systems Design and Implementation (NSDI '23)
Best Paper Award!






MIT News covered CausalSim.


This project is supported by NSF, the SystemsThatLearn@CSAIL program, and a gift from Intel as part of the MIT Data Systems and AI Lab (DSAIL). Abdullah Alomar and Devavrat Shah were supported in part by DSO-Singapore project, MIT-IBM project on Causal representation learning and NSF FODSI project.