4 problems

Conjecture that sub-gradient methods identify non-active strict saddle true solutions
arXiv@arxiv
Sub-gradient identification conjecture. The true solutions identified by the sub-gradient method applied to the $\ell_1$ -loss might be non-active strict saddle points.
Conjectureopenlow-rank-matrix-recovery non-smooth-optimization subgradient-methods
1
Conjecture that highly rank-imbalanced solutions are not local minima in matrix completion
arXiv@arxiv
Rank-imbalance conjecture. Highly rank-imbalanced true solutions are unlikely to emerge as local minima of the loss function.
Conjectureopenlow-rank-matrix-recovery matrix-completion non-smooth-optimization
0
Conjecture that true solutions include non-active strict saddle points in matrix sensing
arXiv@arxiv
Non-active true-solution conjecture. A subset of the true solutions are non-active strict saddle points.
Conjectureopenlow-rank-matrix-recovery matrix-sensing non-smooth-optimization
0
Conjecture that true solutions can be non-active strict saddle points for the ℓ1-loss
arXiv@arxiv
Non-active strict-saddle conjecture. In situations where the true solutions are neither local nor global minima, they may nevertheless manifest as non-active strict saddle points.
Conjectureopenlow-rank-matrix-recovery non-smooth-optimization
1

Conjecture that sub-gradient methods identify non-active strict saddle true solutions