technique

stochastic gradient ascent

techniqueactiveprovisionalstochastic-gradient-ascent-e0807f54·1 events·first seen 21d ago

Aliases: stochastic gradient ascent

Co-occurring entities

Probabilistic Smoothing with Ratio-Monotone Transforms Gaussian kernel smoothing black-box adversarial attacks leave-one-out baseline

More like this (12)

Accelerated Decentralized Stochastic Gradient Descent for Strongly Convex Optimization conjugate gradient gradient accumulation randomized coordinate descent gradient noise scale SDE (Stochastic Differential Equation LR scaling)Integrated Gradients Wasserstein Policy Gradient Gradient-Guided Reward Optimization Private Stochastic Convex Optimization Policy Gradient Methods Gradient Reuse

Recent events (1)

4arXiv · cs.LG·21d ago·source ↗

Probabilistic Smoothing with Ratio-Monotone Transforms for Global Optimization

This paper proposes a generalized probabilistic smoothing framework for global optimization that replaces Gaussian kernels with flexible symmetric unimodal kernels combined with monotonic ratio-based transformations. The authors prove that the smoothed objective preserves the global maximizer and that stationary points concentrate near the true optimum under large amplification, without requiring a decreasing smoothing schedule. Explicit complexity bounds for stochastic gradient ascent are derived, and a leave-one-out baseline is shown to provably reduce variance. Experiments on high-dimensional benchmarks and black-box adversarial attacks demonstrate improved robustness over existing methods.

AI Safety Research stochastic gradient ascent Probabilistic Smoothing with Ratio-Monotone Transforms Gaussian kernel smoothing +2 more