New algorithm offers adversarial resilience for submodular maximization

This research paper introduces a new algorithm called the Spiteful Greedy Swap Poisson Process (SGS-Poisson) for submodular maximization problems with matroid constraints. The algorithm demonstrates adversarial resilience, meaning it maintains its approximation factors even when faced with imperfect value oracles. This resilience allows for the development of full-bandit contextual multi-armed bandit (CMAB) algorithms with improved regret bounds for submodular rewards. AI

IMPACT Introduces theoretical advancements in optimization algorithms relevant to machine learning research.

RANK_REASON The cluster contains a single academic paper detailing a new algorithm and theoretical results. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.AI →

• arXiv
• Hugging Face
• Poisson process
• Spiteful Greedy Swap Poisson Process

• paper
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →