Erik Schmidt, a Senior Scientist at Pandora is going to propose an insight of recommendations and deeper challenges involved with Pandora at the Machine Intelligence Summit. This global tech event will take place in San Francisco on 23rd and 24th of March 2017.
When probed further about his work at Pandora and stuffs that excites him contextual to machine intelligence world, this is all he has to say.
Pandora’s core area of specialization is Machine Listening. To ensure a dominating presence in the sphere of music, our expertise lay in encrypting the musicological content of an audio signal automatically, using sophisticated systems. To work efficiently, these systems need to depend on monitored machine learning and Pandora’s Music Genome Project with a massive 1.5 million canvassed tracks offer the expanding and most elaborate collection in the world, which is just perfect to carry out this task.
The cynosure of Pandora’s recommendation system is a humongous conglomeration, sprawling over 60 recommendation tactics. Although most of our listeners are oblivious but we are continuously adding, subtracting and adjusting these tactics in order to improve our listener’s experience. The best part of our work is when we develop something on the research wing, such as machine-predicted musicological attributes and introduce them into our utilitarian recommendation strategies. It is then that we know how much thumbs up we are getting and within what set of period.
At present, there is an implausible research impetus in the domain of machine learning, encompassing both corporate and academia. While there have been several significant developments in this field recently, the truth is that most of these extremely productive techniques were present for the past few decades. What was lacking until now is the huge amount of data ensemble and computing tools that are now being accumulated by the industry enterprises.
Music is my passion. Compelled by this intense fervour for music, I love exploring the orbits of machine learning to do something more on music discovery. By using productive platforms like Pandora, we can bring out the next big artist and present him right in front of the apt listeners right on time. Advanced tools help in dissecting the audio waveform to match up musicologically with listener’s likes in terms of songs and artists, taking the music experience to an altogether different level. Connecting with the social media to distinguish newbie artists based on terms of popularity is an added advantage.
Frankly speaking, each industry is going to be influenced by machine intelligence in the near future. Machines have already started playing a dominant role in producing goods of daily use. Implementing this transformative technology is surely going to be our biggest challenge, especially the ways in which our society responds will matter a lot.
In the next 5 years, trust me no one has to struggle to find a convenient-cum-effective music discovery tool. If you ask me, conversational AI is surely going to bring in a revolution in this arena of music and machine interaction. Besides, there has been a remarkable progress in the spheres of self-driving cars and complex reasoning tasks and I expect it to continue in the coming years.
To give your career an instant boost, scroll through our machine learning courses in Pune and Delhi NCR at DexLab Analytics.
Interested in a career in Data Analyst?
To learn more about Machine Learning Using Python and Spark – click here.
To learn more about Data Analyst with Advanced excel course – click here.
To learn more about Data Analyst with SAS Course – click here.
To learn more about Data Analyst with R Course – click here.
To learn more about Big Data Course – click here.