Practical Issues in Classification (Training/test error/ ROC curves, k-fold cross-validation)

Hidden Markov models (these notes still require some work near the end...)

Graphical models

1. •Generalization Bounds for Discriminative Sparse Coding Methods, Alexander Gray, coming soon to an arXiv near you
   

2. •Discriminative Sparse Coding for Classification and Regression, Alexander Gray, The Learning Workshop, 2011. Oral (Note that the abstract isn’t precisely representative of the oral)
   

3. •Generative and latent mean map kernels, Alexander Gray, arXiv 1005.0188, 2010
   

4. •Recognizing sign language from brain imaging, Thad Starner, Melody Moore Jackson, Karolyn Babalola, and George Andrew James. International Conference on Pattern Recognition, 2010
   

5. •  Longer, earlier version: Recognizing sign language from brain imaging, GVU Technical Report GIT-GVU-09-06, 2009.
   

6. •Optimal control strategies for an SSVEP-based brain-computer interface, Sadhir Hussain and Melody Moore Jackson, 2010 (preprint forthcoming)
   

7. •FuncICA for time series pattern discovery, Alexander Gray, SIAM Data Mining, 2009. Selected for oral presentation
   

8. •Estimating neural signal dependence using kernels, Alexander Gray, Thad Starner, and Melody Moore Jackson. NIPS Workshop on Statistical Analysis and Modeling of Response Dependencies in Neural Populations, 2008

I'm a fifth year PhD student in machine learning at the College of Computing at Georgia Tech. My advisor is Alexander Gray (FASTlab). In the summer of 2010 I enjoyed a research internship at Microsoft Research with Alice Zheng.

Research Interests: Discriminative embeddings. Learning theory for embeddings, manifolds, and kernel machines. Statistical machine learning algorithms for structured data. Active learning. Information geometry.

Now: 

  Riemannian Geometry (Do Carmo)

(some) Knowledge extracted:

  Introduction to Smooth Manifolds (Lee)

  Entropy, Compactness, and the Approximation of Operators (Carl and Stephani)

  Learning Theory: An Approximation Theory Viewpoint (Cucker and Zhou)

  Support Vector Machines (Steinwart and Christmann)

  Combinatorial Methods in Density Estimation (Devroye and Lugosi)

  Probabilistic Theory of Pattern Recognition (Devroye, Györfi, Lugosi)

In my dreams:

  Methods of Information Geometry (Amari)

  The Wind-Up Bird Chronicle (Murakami)

  Prediction, Learning, and Games (Cesa-Bianchi and Lugosi)

Panita Thai Kitchen - 1

Spoon - 1

Tamarind Seed - 1

Tenth - ?

Bangkok Thai Restaurant - ?

Royal Orchid - 0

Top Spice - 0

Surin - 0

Current:

  Geometry and topology II, Advanced classical probability theory

Past:

  Geometry and Topology I, Introduction to graph theory, Information theory

  Machine learning theory, Probabilistic graphical models, Combinatorial methods in density estimation

  Design and analysis of algorithms, Spectral algorithms

  Linear optimization, Nonlinear optimization

Nishant Mehta

niche @t |c|c.ga|tech.ed|u

College of Computing

Georgia Institute of Technology

266 Fifth St NW

Atlanta GA 30332

cv

Photos

Learning Machinations