Math Meets Music To Aid The "Piki" Listener
An Interview With Piki Founder & Cornell Mathematician Sasha Stoikov
German-American Sasha Stoikov grew up in Switzerland with a passion for numbers and musical notes. Stoikov’s love for mathematics stemmed from two things he says, “I pretty much always knew that I was a bit better at math than all the other things… I think what I liked about [it] was that there was an ultimate truth. That’s comforting for me, building opinions based on facts.” Mr. Stoikov moved to the United States, Boston specifically, for college–where he caught the probability bug. From stock market predictions to music recommender systems, Stoikov’s fascination with probability has stuck with him throughout the entirety of his academic and professional career. Stoikov currently works as a Senior Research Associate at Cornell Financial Engineering Manhattan.
While Stoikov has always had a love for mathematics, music has also always been a passion of his. He played guitar as a kid but stopped playing for a few years until he attended UT Austin for grad school. Life in Austin is centered around live music and supporting middle class musicians who live from their live shows. So naturally, one of Stoikov’s projects, a mobile app called Piki, aims to get people off of their streaming devices and out to venues to support local musicians. Piki presents users with 30 second clips of suggested songs based on the user's favorite songs and then provides a list of nearby shows and how to get tickets. Stoikov says “It is not a passive listening experience.” Users “like” and “dislike” songs, and each choice is recorded and fed into the program’s algorithm (written by Stoikov himself), and new suggestions are generated based on the previous data.
The idea for Piki came to be while Stoikov was working at Cantor Fitzgerald as the VP of algorithmic trading. Around this time, Stoikov found an anonymous music data set that spoke to him, and he began to play around with the numbers. The data set consisted of over a million anonymous users and their 100 most listened to songs. Stoikov figured, “If I make a list of songs that I really like that are all over the place, from Beethoven to Daft Punk passing through Patti Smith and Childish Gambino...and search amongst the list of people and find the closest match, then technically I should like most of their music as well.” Stoikov claims it’s all about “building a theory, refining and fine turning it, that's when the modeling begins,” and that is exactly what he did. Following the transitive property, if person A likes 75% of the same songs as person B, then the other 25% percent of person B’s music that is undiscovered by person A, should be a positive match. Stoikov says that one of the most exciting features is that “you might think you're gonna match up with people who only like one type of rock and roll, but most people are eclectic and like multiple genres, so you get some matches that are outside of the obvious.” These unexpected matches are what allow for true new music discovery and are what lead to the support of local artists and music venues.
Although inspired by Austin’s local musicians, Piki is designed specifically for exploring New York City’s live music scene. The name Piki NYC was chosen, Stoikov claims, because “one thing we found out about New Yorkers who love music is that they are picky.” Stoikov is excited for the future of Piki because “the app itself will generate data, whereas up to now, I’ve been reliant on other people’s data.” Stoikov takes issue with the idea of echo chambers where users are just presented with news, information, and or movies similar to the ones they’ve already seen. Stoikov wants to make the unseen connections, presenting users with new content that they wouldn’t have easily found on their own.
Stoikov’s projects cover a vast range of topics from market microstructure–where he studied limit order books–to music recommender systems–in which he analyzed trends in users most listened to tracks. The common thread throughout all of his work, Stoikov says, is that he “loves working with very large data sets where some form of intelligence can be weaned out of it.” In 2007, a banker shared a financial order book data set with Stoikov that was not available to the public. This data is what led to his research in market microstructure, an area that has grown in importance with the emergence of electronic financial markets. “It has always started with finding a data set that’s interesting... if you have a model that’s great, but if you can’t fit it to data it’s just a toy.” Stoikov’s relationship with mathematics and his work has shifted throughout the years, as he states, “When I first started college I was just into the beauty of pure math, but now I have the opinion that fitting theoretical ideas to data is the most exciting part of science.”
Looking towards the future, Stoikov says that he is motivated right now by “technology that makes people happy not miserable. I sort of find that...there's a lot of people who spend hours on their phones and constantly feel that they are missing out on things or are unhappy.” He believes that “technology is at its best when it helps you find things to do in the real world,” and looks forward to technology disappearing from the forefront of human social interactions and rather be used to enable people to do things in the real world. Stoikov wants to connect people at opposite ends of “the cloud”. The dirty secret of AI, Stoikov claims, “is that behind what the machine tells you is just an average of a lot of people’s opinions...a robot saying you will like this music is a lot less exciting than someone saying ‘let’s go to a show.’” Stoikov “would like to see more human connection in the real world outside of that technological space,” and so far he has used his mathematical and musical background to do just that.
To learn more about Stoikov’s app Piki, visit https://www.piki.nyc/
Piki can also be downloaded from the Apple App Store
Interview with NASA Astronaut Scott Kelly: An American Hero
13 Questions With General David Petraeus
Why Choose Machine Learning Over A Traditional Financial Advisor?
Deep Fakes: Terror In A Data Driven World
Nuclear Submarines: A 7,000 Lb Swiss Watch
Interview with the Inventor of Amazon's Alexa
Ai Can Write Its Own Computer Program
On Black Holes: Gateway to Another Dimension, or Ghosts of Stars’ Pasts?
Tesla's Augmented Reality & Virtual Reality Founder Tyler Lindell Opens Up
Supersonic Travel: The Future of Aviation
Lockheed Martin Confirms the SR-72 – Son of Blackbird Will Reach Anywhere in the World in One Hour
Conversation with Bloomberg's Former Head of Trading on the Future of Humans in the Trading Industry
Shedding Light on Dark Matter: Using Machine Learning to Unravel Physics’ Hardest Questions
Aquaponics: How Advanced Technology Grows Vegetables In The Desert
The Perfect Beer: Using Artificial Intelligence to Personalize Goods
The World Cup Does Not Have a Lasting Positive Impact on Hosting Countries
The Implications of Machine Learning on Condensed Matter Physics & Quantum Computing
Written by Grace Kelman, Edited by Alexander Fleiss