### 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.