Guangjian Zhang

Randomization Is Necessary for Learning from Symmetric Equivalence Queries

We resolve the dimension-only derandomization question for general symmetric equivalence queries posed in "Learning from Equivalence Queries, Revisited" at COLT 2026. Prior work gave a randomized proper learner with expected query complexity O(d) for every finite class of Littles...

The Price of Adaptivity in Bandit Multiclass Classification: Minimal Separation and Sharp Tower Asymptotics

We resolve the existence component of a foundational open problem posed at NeurIPS 2024 concerning the price of adversarial adaptivity in bandit multiclass classification. The optimal expected mistake bound against an adaptive adversary is at most a multiplicative O(k) above the...

Quadratic Sample Complexity for Strictly Replicable Realizable Learning with Unknown Marginals

Under the quantifier "there exists A for all D", where the realizable distribution and its marginal are unknown, we determine the finite-class sample-complexity exponent for strict replicability: two independent samples and one shared seed must produce identical functions. For ep...

Pointwise Replicable Agnostic Learning: The Additive Term Is Unnecessary and the Fixed-Confidence Order Is Exact

Hopkins, Impagliazzo, and Ye (arXiv:2510.20200) introduced $\rho$-pointwise replicable learning and proved that every class of VC dimension $d$ admits a pointwise replicable agnostic $(\alpha,\beta)$-learner with $\widetilde O(d/(\rho^2\alpha^2)+1/(\rho^2\alpha^4))$ samples, aski...

Pointwise Replicable Agnostic Learning: The Additive Term Is Unnecessary and the Fixed-Confidence Order Is Exact

Hopkins, Impagliazzo, and Ye (arXiv:2510.20200) introduced $\rho$-pointwise replicable learning and proved that every class of VC dimension $d$ admits a pointwise replicable agnostic $(\alpha,\beta)$-learner with $\widetilde O(d/(\rho^2\alpha^2)+1/(\rho^2\alpha^4))$ samples, aski...

Quadratic Sample Complexity for Strictly Replicable Realizable Learning with Unknown Marginals

Under the quantifier "there exists A for all D", where the realizable distribution and its marginal are unknown, we determine the finite-class sample-complexity exponent for strict replicability: two independent samples and one shared seed must produce identical functions. For ep...