## Nonnegative Matrix Factorization

Discussion: When does well-posedness lead to better algorithms?

Balcan, M., A. Blum, et al.
“Clustering under Approximation Stability.” (PDF)
*Journal of the ACM* (2013).

## Tensor Decompositions

Discussion: When do algorithms rely (too much) on a distributional model?

Feige, U., and J. Kilian. “
Heuristics for Semirandom Graph Problems
.” *Journal of Computing and System Sciences* 63, no. 4 (2001): 639–71.

## Sparse Coding

Discussion: When does belief propagation (provably) work?

Geman, S., and D. Geman. “
Stochastic Relaxation, Gibbs Distributions, and the Bayesian Restoration of Images
.” *Pattern Analysis and Machine Intelligence* (1984).

## Learning Mixture Models

Discussion: Is nature an adversary? And if not, how can we model and exploit that?

Bhaskara, A., M. Charikar, et al. “
Smoothed Analysis of Tensor Decompositions
.” *Symposium on Theory of Computing* (2014).

## Linear Inverse Problems

Discussion: Do we have enough average-case assumptions?

Berthet, Q., and P. Rigollet. “
Computational Lower Bounds for Sparse PCA
.” *Conference on Learning Theory* (2013).

Chandrasekaran, V., and M. Jordan. “
Computational and Statistical Tradeoffs via Convex Relaxation
.” *Proceedings of the National Academy of Sciences of the United States of America* 110, no. 13 (2013): E1181–90.