AI Paper Atlas

Research map / Learning theory and algorithms / Combinatorial and search optimization

A Penalty Approach For Differentiation Through Black-box Quadratic Programming Solvers

Yuxuan Linghu, Zhiyuan Liu, Qi Deng

ICML 2026 · Learning theory and algorithms › Combinatorial and search optimization · approach: Predict-then-optimize learning · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
replace KKT-based backward pass (dQP) with smoothed-penalty primal SPD solve (dXPP)probability simplex projection, size 1000000 · backward runtimebackward time (ms): 950.64 -> 226.21 (4.2x speedup)Improved
replace dQP backward with dXPPchain projection, size 1000000 · backward runtimebackward time (ms): 3071.18 -> 333.42 (9.2x speedup)Improved
replace dQP with dXPP in dense mode multi-period portfolio optimizationhorizon H=200 · per-epoch backward runtimebackward time (ms): 39105.75 -> 113.97Improved
dXPP versus OptNet on multi-period portfoliohorizon H=200 · per-epoch backward runtimebackward time (ms): 605.89 -> 113.97 (5.3x)Improved
dXPP versus OptNet at very small problem sizesimplex projection, size 20 · backward runtimebackward time (ms): OptNet 0.54 vs dXPP 0.72Hurt
use dXPP gradients instead of dQP gradients on random QPsrandom strictly convex QPs · gradient accuracyrelative gradient difference: 1.91 × 10−7 (10x5) to 5.14 × 10−4 (5000x2000)No clear effect
choose intermediate smoothing scale rather than very large or very small (ablation)dXPP · random QPs gradient accuracyrelative gradient difference: 5000x2000: 3.18×10−6 (1e-5), 2.62×10−7 (1e-7), 1.63×10−5 (1e-9)Mixed
prune inactive-constraint terms (omit E_delta and F_delta) in the backward pass (ablation)dXPP · performancegradient qualityNo clear effect
use dXPP forward pass with Gurobi instead of dQP in dense modemulti-period portfolio · total runtimetotal time (ms): 40066.11 -> 977.98 at H=200Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy