MIT researchers may have found a way to kill video buffering

If this works, video buffering could be reduced by as much as 30%

One thing that puts people off is video buffering. Watching a YouTube video only for it to get stuck in the middle is one of the most frustrating things there is on the internet.

Not everyone has a fast internet connection and has likely suffered such problems.

However, this might change for the better as MIT researchers might have come up with a solution to this.

With the help of machine learning, the Pensieve system figures out the optimal algorithm to use for delivering video keeping in mind the best possible resolution while avoiding buffering breaks, no matter your internet speed.



Currently, the systems make a trade-off by compromising on video quality and the amount of re-buffering it needs to view the clip, however, with the proposed tools, systems will use the most suitable AI depending on the situation to reduce buffering by as much as 30%.

“Studies show that users abandon video sessions if the quality is too low, leading to major losses in ad revenue for content providers,” according to MIT Professor Mohammad Alizadeh.



In the experimental phase, teams found that Pensieve could stream video with 10 to 30 per cent less rebuffering at levels that users rated 10 to 25 percent higher on key “quality of experience” (QoE) metrics.

The researchers will present their paper at the upcoming SIGCOMM Conference in Los Angeles.