9 problems

Let $y^1,y^2,y^3,y^4$ be interpolated points, and write $\overrightarrow{y^i y^j}$ for the vector from $y^i$ to $y^j$. A point is called bad under $\Lambda$-poisedness when it sati…

Let $\mathcal{Y}$ be the set of experimental points, and let $y^1,y^2,y^3\in\mathcal{Y}$. A point is called bad under $\Lambda$-poisedness when it satisfies the paper's bad-point c…

Assume $f obreaktext{ belongs to } C^{1,1}nu(mathbb{R}^n)$. Let $hat{f}$ be the linear function that interpolates $f$ at any set of $n+1$ affinely independent vectors…

Shamir's conjecture. The rate

Conjecture on derivative-free error accumulation. In the derivative-free setting, it is impossible to obtain a slower rate of error accumulation than the rate achieved by the autho…

Bubeck–Kleinberg–Slivkins conjecture. Their algorithm has complexity

Quadratic interpolation convergence conjecture. One might conjecture that the resulting algorithm converges similarly to a second-order method, that is, quadratically.

Trust-region radius convergence conjecture. Almost surely,

Consider the derivative-free stochastic convex optimization setting described above, including the quadratic strongly convex objectives and noisy function-value queries. Gradient-b…