Parvez Ahammad: SpeedPerception

San Francisco & Silicon Valley Web Performance Group
San Francisco & Silicon Valley Web Performance Group
Public group

Instart Logic

450 Lambert Avenue · Palo Alto, CA

How to find us

Caltrain: California Avenue

Location image of event venue


The Halloween / October Web Performance Meetup is hosted at Instart Logic’s Palo Alto office on Thursday, October 27th, 2016 at 6:30PM. Doors open at 6:00PM. Food and drinks will be provided by our sponsor Instart Logic (thank you!).

Please only RSVP if you are fairly confident you will be attending, and change your status to "No" if you realize you can no longer attend.

Halloween costumes are highly encouraged! The more zombies the merrier.

About this talk:

Parvez Ahammad will cover the first results from the SpeedPerception ( challenge, with over 5,000 sessions completed and over 50,000 valid data points.

The first phase of the SpeedPerception study focused on three hypotheses, of which two were confirmed:

✔ No single metric can explain human choices with 90%+ accuracy
✗ Visual metrics will perform better than non-visual/network metrics
✔ Users will not wait until “visual complete” to make a determination

For those of you unaware, SpeedPerception is a free, open-source, benchmark dataset of how people perceive above-the-fold rendering and webpage loading processes, which can be used to better understand the perceptual aspects of end-user web experience. The benchmark we’ve posted on Github ( can provide a quantitative basis to compare different algorithms. Our hope is that this data will spur computer scientists and web performance engineers to make progress quantifying perceived web performance.

• Dataset, study and results:
• Challenge Site:

Parvez will be presenting on the initial findings. The presentation of the results will be followed by a question and answer period during which time you will be encouraged to provide feedback on the study, the results, and any suggestions for next steps.

In the meantime, if you want to analyze the data yourself and test your own hypotheses, the data and code are all available on Github. Please do share any results and conclusions with us.

Thanks to Parvez Ahammad, Clark Gao, Prasenjit Dey, Estelle Weyl, Pat Meenan and the entire web performance community for helping make this study a reality.

About the speaker:

Parvez Ahammad, PhD: With expertise in computer vision, machine learning, statistics, and signal processing, Parvez Ahammad leads the data science and machine learning group at Instart Logic Inc. Parvez has worked on a wide array of problems in domains of web application delivery, neuroscience, bioinformatics, and heterogeneous sensor networks. He is the creator of novel algorithmic technologies such as SmartVision at Instart Logic, OpSIN and Salient Watershed at HHMI–Janelia, and fast video activity recognition at UC-Berkeley.


18:00 :: Check-in & Mingle
18:30 :: Announcements and Introductions
18:45 :: Presentation
19:30 :: Questions and Answer & Open Form
20:00 :: Networking & Discussion


Instart Logic
450 Lambert Avenue
Palo Alto CA 94306

Caltrain: California Ave stop.


Instart Logic ( Endpoint-aware application delivery solution that improves applications with an intelligent architecture optimizing and securing website elements based on a user’s device, browser, and network conditions.