pith. sign in

arxiv: cs/0408007 · v1 · submitted 2004-08-02 · 💻 cs.LG · cs.CC

Online convex optimization in the bandit setting: gradient descent without a gradient

classification 💻 cs.LG cs.CC
keywords gradientpointfunctionsettingchosenconvexonlinesingle
0
0 comments X
read the original abstract

We consider a the general online convex optimization framework introduced by Zinkevich. In this setting, there is a sequence of convex functions. Each period, we must choose a signle point (from some feasible set) and pay a cost equal to the value of the next function on our chosen point. Zinkevich shows that, if the each function is revealed after the choice is made, then one can achieve vanishingly small regret relative the best single decision chosen in hindsight. We extend this to the bandit setting where we do not find out the entire functions but rather just their value at our chosen point. We show how to get vanishingly small regret in this setting. Our approach uses a simple approximation of the gradient that is computed from evaluating a function at a single (random) point. We show that this estimate is sufficient to mimic Zinkevich's gradient descent online analysis, with access to the gradient (only being able to evaluate the function at a single point).

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 14 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Prudent-Banker: No Extra Fees for Baseline Safety in Adversarial Bandits With and Without Delays

    cs.LG 2026-05 unverdicted novelty 7.0

    Prudent-Banker achieves pseudo-regret Õ(√T + √D) and Õ(1) regret vs. safe comparator in adversarial bandits both with and without delays, matching new lower bounds up to logs.

  2. Bandit Convex Optimization with Gradient Prediction Adaptivity

    cs.LG 2026-05 unverdicted novelty 7.0

    TP-VR-OPT achieves O(√(d E[S_T])) prediction-adaptive regret in two-point bandit convex optimization, with a matching Ω(√E[S_T]) lower bound up to √d, while single-point feedback cannot benefit from predictions.

  3. Online Conformal Prediction with Corrupted Feedback

    cs.LG 2026-05 unverdicted novelty 7.0

    Develops robust online conformal prediction schemes that provide explicit miscoverage guarantees under feedback modeled as arbitrary binary flips or bounded-memory errors.

  4. Select-then-differentiate: Solving Bilevel Optimization with Manifold Lower-level Solution Sets

    math.OC 2026-05 unverdicted novelty 7.0

    Optimistic bilevel optimization with manifold lower-level minimizers is differentiable if the optimistic selection is unique, yielding a pseudoinverse hyper-gradient and a convergent HG-MS algorithm whose rate depends...

  5. NePPO: Near-Potential Policy Optimization for General-Sum Multi-Agent Reinforcement Learning

    cs.LG 2026-03 unverdicted novelty 7.0

    NePPO learns a player-independent potential function via a novel objective whose minimization yields an approximate Nash equilibrium for general-sum multi-agent games.

  6. Baryon-antibaryon photoproduction cross sections off the proton

    hep-ex 2025-10 unverdicted novelty 7.0

    GlueX measured total and differential cross sections for baryon-antibaryon photoproduction, finding forward-peaked distributions consistent with t-channel exchange, a phenomenological double-exchange model fit, and no...

  7. Stochastic Zeroth-Order Optimization Under Heavy-Tailed Noise

    math.OC 2026-05 unverdicted novelty 6.0

    RSC-ZO achieves high-probability ε-stationary points for stochastic ZO optimization under weak-L_p heavy-tailed noise with Õ(d^{p/2(p-1)} ε^{-(3p-2)/(p-1)}) function queries.

  8. Global Convergence of Sampling-Based Nonconvex Optimization through Diffusion-Style Smoothing

    cs.LG 2026-05 unverdicted novelty 6.0

    Recasts sampling-based nonconvex optimization as smoothed gradient descent to obtain non-asymptotic convergence guarantees and introduces the DIDA annealed algorithm that converges to the global optimum.

  9. Ensemble Distributionally Robust Bayesian Optimisation

    cs.LG 2026-05 unverdicted novelty 6.0

    A tractable ensemble distributionally robust Bayesian optimization method achieves improved sublinear regret bounds under context uncertainty.

  10. OGPO: Sample Efficient Full-Finetuning of Generative Control Policies

    cs.LG 2026-05 unverdicted novelty 6.0

    OGPO is a sample-efficient off-policy method for full finetuning of generative control policies that reaches SOTA on robotic manipulation tasks and can recover from poor behavior-cloning initializations without expert data.

  11. Policy Optimization for Unknown Systems using Differentiable Model Predictive Control

    eess.SY 2025-11 unverdicted novelty 6.0

    Hybrid differentiable-plus-zeroth-order policy optimization for MPC yields faster transients than pure data-driven methods while retaining convergence guarantees under model mismatch, demonstrated on a 12D quadcopter.

  12. Stochastic Regret Guarantees for Online Zeroth- and First-Order Bilevel Optimization

    cs.LG 2025-11 unverdicted novelty 6.0

    Introduces a novel search direction enabling sublinear stochastic bilevel regret guarantees for first- and zeroth-order online bilevel optimization algorithms without relying on window smoothing.

  13. Inference-Time Scaling for Diffusion Models beyond Scaling Denoising Steps

    cs.CV 2025-01 conditional novelty 6.0

    Diffusion models improve generation quality via inference-time search over noise candidates guided by verifiers and algorithms, yielding gains beyond denoising step scaling on class- and text-conditioned benchmarks.

  14. Position: Zeroth-Order Optimization in Deep Learning Is Underexplored, Not Underpowered

    cs.LG 2026-05 unverdicted novelty 5.0

    Zeroth-order optimization is underexplored rather than underpowered in deep learning, with limitations stemming from full-space designs that can be addressed via subspace, spectral, and systems-aware approaches.