Abstract Given a set of driver logs collected from Unity simulator, our task is to predict driver’s aggressive events, including overspending, driving too close, lane switch to left, and lane switch to right in the near future (10 secs). By achieving this goal in real time with the assumption of effective communication between cars, we
Category: Past Researches
SB Hackathon, University of California, Santa Barbara
Abstract During this Hackathon, we are working on an advertisement related machine learning topic: predict the possibility of conversion (click on an advertisement and buy the item) given logs of past users’ behaviors. It is unlike usual collaborative filtering task, where only interactions, ex: ratings, are provided; on the contrary, many contents of users and
Continue Reading “SB Hackathon, University of California, Santa Barbara”
Big Data TSMC Semi-conductor Competition: Big Data Analytics for Semiconductor Manufacturing
Abstract Semiconductor is the foundation of modern embedded systems. In this high computation power era, the circuit probing value of semiconductor production is widely concerned. In this series of competition held by TSMC, we have more than 100,000 features by hand. Thus, the feature dimension reduction is a very key procedure for us. On top
College Student Research Scholarship: Real-time classification with missing data given models
Abstract With the explosion amount of information caused by Internet expansion, machine learning is getting more and more popular recently. However, real world application of machine learning theory is still very challenging and hard to achieve. One of the main difficulties stops the direct application is the issue called data loss. No matter data loss
Recent Comments