Over the years, many different traits have tried to define the human character, but at the same time, none have done a better job than that tendency of ours to grow on a consistent basis. This inclination towards improving, no matter the situation, has already fetched the world some huge milestones, with technology emerging as quite a major member of the group. The reason why we hold technology in such a high regard is, by and large, predicated upon its skill-set, which ushered us towards a reality that nobody could have ever imagined otherwise. Nevertheless, if we look beyond the surface for a second, it will become clear how the whole runner was also very much inspired from the way we applied those skills across a real world environment. The latter component, in fact, did a lot to give the creation a spectrum-wide presence, and as a result, kickstart a tech revolution. Of course, this revolution then went on to scale up the human experience through some outright unique avenues, but even after achieving a feat so notable, technology will somehow continue to bring forth the right goods. The same has turned more and more evident in recent times, and assuming one new discovery ends up with the desired impact, it will only put that trend on a higher pedestal moving forward.

The researching team at Massachusetts Institute of Technology has successfully developed a new tool called CausalSim, which is designed to eliminate the bias that has long plagued the process of simulating complex systems. To understand the significance of such a development, we must start from looking into the current method’s shortcomings. You see, considering it’s practically impossible for a researcher to capture every detail of a complex system in simulation, they focus on the next best option i.e. collecting small amounts of  real data that can be replayed while simulating the study’s primary components. This is what we call a trace-driven simulation, and the main problem with it is how the method, at times, can produce substantially biased outcomes, something that makes the whole research inaccurate and pointless. Fortunately, the all-new CausalSim tool promises to change the stated reality. According to certain reports, the tool initiates the show by taking into account the collected trace data. Once this step is done, it estimates the underlying functions that produced the data. Notably enough, the tool makes a point to inform researchers on those exact same underlying conditions that a user experienced, and how a new algorithm can meaningfully alter the outcome.

This promised efficacy was on full-display when MIT researchers used CausalSim to design an improved bitrate adaptation algorithm. Going by the published results, the tool empowered them to select one variant which gave a stall rate nearly 1.4 times lower than a well-accepted competing algorithm, and mind you, it did the job while achieving, more or less, the same video quality. 

When quizzed regarding the new tool, Arash Nasr-Esfahany, an electrical engineering and computer science (EECS) graduate student and co-lead author of a paper on this new technique, offered the following explanation:

“Data are not the only thing that matter. The story behind how the data are generated and collected is also important. If you want to answer a counterfactual question, you need to know the underlying data generation story so you only intervene on those things that you really want to simulate,”

In contrast to their experience with CausalSim, when the researchers did a similar study using the typical trace-driven simulator, they observed completely opposite results, as the simulator predicted the established variant should cause a stall rate that was nearly 1.3 times higher. Caught between conflicting predictions, the team then tested the algorithm on a real-world video streaming platform where it was confirmed that CausalSim was correct. 

“The gains we were getting in the new variant were very close to CausalSim’s prediction, while the expert simulator was way off. This is really exciting because this expert-designed simulator has been used in research for the past decade. If CausalSim can so clearly be better than this, who knows what we can do with it?” said Pouya Hamadanian, a researcher involved in the study.

Surely, CausalSim can improve your video streaming platforms, as well as those expansive data processing systems, but the researcher’s wider vision is to use it wherever randomized control trial data is not available or where it is especially difficult to recover the causal dynamics of the system.