Tagged articles
multi-armed bandit
11 articles · Page 1 of 1
FlashSpec implements speculative decoding using GPU-native Triton kernel verification and online bandit-based draft model selection, sharing six practical lessons on specification-first development, hidden temperature bugs, cross-platform packaging pitfalls, kernel performance trade-offs, property-based testing value, and adaptive algorithm prerequisites.
CI/CDLLM inferenceThompson sampling
0 likes · 15 min read

Multi‑armed bandit models illustrate the core exploration‑exploitation dilemma in reinforcement learning, covering greedy, ε‑greedy, and optimistic‑initial‑value strategies, as well as sample‑average and incremental Q‑value estimation methods with practical examples and visual illustrations.
Q-value estimationReinforcement Learningexploration vs exploitation
0 likes · 15 min read

This article presents a self‑adaptive genetic algorithm framework enhanced with multi‑armed bandit techniques to tackle pricing volatility in the freight industry, detailing its design, business challenges, experimental validation, and future integration with large language models for smarter dynamic pricing.
Dynamic Pricingadaptive optimizationfreight logistics
0 likes · 22 min read

This article explains the principles of A/B testing and the ε‑greedy multi‑armed bandit algorithm, illustrates their practical use in e‑commerce recommendation optimization, and draws broader life lessons about balancing exploration and exploitation for better personal and professional decisions.
A/B testingexploration vs exploitationgreedy
0 likes · 6 min read


58 TechDec 28, 2021 · Artificial Intelligence
This talk explains how 58.com tackles the cold‑start and interest‑divergence problems of its massive blue‑collar job recruitment platform by modeling the recommendation process as a reinforcement‑learning task, detailing the use of multi‑armed bandit, contextual bandit, and linear‑UCB algorithms, offline evaluation pipelines, online deployment, and observed performance gains.
Contextual BanditReinforcement Learningcold start
0 likes · 25 min read

This talk explains how 58.com’s massive blue‑collar recruitment platform uses reinforcement‑learning techniques—including multi‑armed bandits, contextual MAB, and linear UCB—to address cold‑start and interest‑divergence challenges, describes the system architecture, offline evaluation, online deployment, and reports an 8% uplift in new‑user conversion.
Recommendation SystemReinforcement Learningcold start
0 likes · 26 min read

This article explains how reinforcement learning, with its focus on maximizing long‑term reward, can improve recommendation system mixing by covering basic RL concepts, differences from supervised learning, multi‑armed bandit approaches, practical OpenAI Gym experiments, new AUC metrics, online gains, and advanced model optimizations.
OpenAI GymQ-LearningReinforcement Learning
0 likes · 10 min read

This article describes how iQIYI’s technical product team designed and implemented an AI‑driven personalized poster generation and distribution pipeline that automatically creates, ranks, and serves customized video posters, improving click‑through rates across TV and mobile platforms.
AIPersonalized RecommendationPoster Generation
0 likes · 11 min read

This article examines iQIYI's end‑to‑end personalized poster production and distribution system, detailing AI‑driven image cropping, smart frame extraction, feature extraction, multi‑armed bandit ranking, and online experiments that together significantly increased poster click‑through rates on TV and mobile platforms.
AI poster generationPersonalized RecommendationVideo platform
0 likes · 12 min read

This article introduces the PH‑MAB mechanism that combines public and private transaction histories to improve multi‑armed bandit‑based recommendation systems, explains its truthful mechanism‑design foundation, and shows how it reduces regret and boosts platform revenue compared to traditional epsilon‑greedy approaches.
e-commercemechanism designmulti-armed bandit
0 likes · 6 min read

This article explains how a simple 20‑line multi‑armed bandit implementation can replace traditional A/B testing by continuously balancing exploration and exploitation to automatically discover the most effective UI variant, reducing manual analysis and improving conversion rates.
A/B testingExploitationexploration
0 likes · 8 min read
