18-797/11-755 Poster Presentation

  • Gates&Hillman Centers
  • 6th Floor Atrium
Project Presentations

The annual poster session for 18797/11755 Machine Learning for Signal Processing will be held in the hallway down from the elevator, leading into the bridge between Gates&Hillman.  The projects focus on the application of machine learning techniques to the analysis of signals.

There will be 12 projects, including among others, an attempt at identifying minerals from hyperspectral imagery, and a team who created a map of Porter Hall using only signals from the sensors of smartphones. A couple of brave teams attempted to recreate wavenet, and one student tried to create a face database by applying his award-winning face authentication system to TV videos. We depend on you to grade the projects, so please join in help evaluate the following projects:

  1. Unsupervised Learning of Minerals from Orbital Hyperspectral Imagery Modeling Patient
  2. Mortality from Clinical Text by Combining Topic Modeling and Ontological Feature Learning with Group Regularization
  3. Global Conditional WaveNet
  4. Hand-motion-control user interface using EMG Signal with Kinect Sensor
  5. Deep Generative Model for Voice
  6. Unsupervised Learning for indoor localization and mapping
  7. Landmark Based Audio Fingerprinting and its Application in Audio Retrieval and Duplicate Video identification
  8. Dis-aggregating electricity load using Machine Learning Techniques
  9. Audio-based Depression Detector
  10. Audio Fingerprinting for Timeline Reconstruction
  11. Intoxicated Speech Detection with Neural Network and Gaussian Mixture Model supervectors
  12. Labelling TV videos to create a face dataset

Faculty Host: Bhiksha Raj

For More Information, Please Contact: