Tensor Methods
MIT, UC Irvine
Guaranteed machine learning using tensor methods, with examples of learning probabilistic models and representations.
Taught by: Anima Anandkumar, UC Irvine
Slides:
Additional Resources:
- Introduction to Tensor Methods resource page
- Nonconvex Optimization: Challenges and Recent Successes, tutorial presented by Anima Anandkumar at the 2016 International Conference on Machine Learning
- Mhaskar, H., Liao, Q. & Poggio, T. (2016) Learning functions: when is deep better than shallow, CBMM Memo 45, https://arxiv.org/pdf/1603.00988v4.pdf.
- MATLAB code to implement the power method for tensor decomposition
- C++ code for topic modeling with the method of moments
- GitHub: Spark code to implement the spectral learning method for learning a Latent Dirichlet Allocation (LDA) topic model