8 papers
A Unified Framework for Iterate Convergence of Bregman Proximal Methods
He Chen, Jiaming Fan, Anthony Man-Cho So
Iterate convergence of Bregman proximal methods (BPMs) has long remained open, especially for nonconvex objectives. Recently, \citet{chen2026skl} made progress by establishing iter…
On the Iterate Convergence of Bregman Projected Gradient Method
He Chen, Anthony Man-Cho So
The iterate convergence of \textit{Bregman projected gradient method} (BPGM) has remained a long-standing open problem, especially for the widely adopted Shannon entropy kernel. Ex…
Accelerated Price Adjustment for Fisher Markets with Exact Recovery of Competitive Equilibrium
He Chen, Chonghe Jiang, Anthony Man-Cho So
The canonical price-adjustment process, tâtonnement, typically fails to converge to the exact competitive equilibrium (CE) and requires a high iteration complexity of $\tilde{\math…
Set Smoothness Unlocks Clarke Hyper-stationarity in Bilevel Optimization
He Chen, Jiajin Li, Anthony Man-Cho So
Solving bilevel optimization (BLO) problems to global optimality is generally intractable. A common surrogate is to compute a hyper-stationary point -- a stationary point of the hy…
Probe-Free Low-Rank Activation Intervention
Chonghe Jiang, Bao Nguyen, Anthony Man-Cho So +1
Language models (LMs) can produce texts that appear accurate and coherent but contain untruthful or toxic content. Inference-time interventions that edit the hidden activations hav…
Computing Competitive Equilibrium for Chores: Linear Convergence and Lightweight Iteration
He Chen, Chonghe Jiang, Anthony Man-Cho So
Competitive equilibrium (CE) for chores has recently attracted significant attention, with many algorithms proposed to approximately compute it. However, existing algorithms either…
Spurious Stationarity and Hardness Results for Bregman Proximal-Type Algorithms
He Chen, Jiajin Li, Anthony Man-Cho So
Bregman proximal-type algorithms (BPs), such as mirror descent, have become popular tools in machine learning and data science for exploiting problem structures through non-Euclide…
Lower-level Duality Based Reformulation and Majorization Minimization Algorithm for Hyperparameter Optimization
He Chen, Haochen Xu, Rujun Jiang +1
Hyperparameter tuning is an important task of machine learning, which can be formulated as a bilevel program (BLP). However, most existing algorithms are not applicable for BLP wit…