---
id: ai-for-online-advertising
title: "AI for Online Advertising: Real-Time Bidding, CTR Prediction, and Programmatic Ads"
schema_type: article
category: ai
language: en
confidence: high
last_verified: "2026-05-24"
created_date: "2026-05-24"
generation_method: ai_assisted
ai_models:
  - claude-4.5-sonnet
derived_from_human_seed: true
conflict_of_interest: none_declared
is_live_document: false
data_period: static
completeness: 0.85
atomic_facts:
  - id: af-ai-for-online-advertising-1
    statement: >-
      Real-time bidding (RTB) processes 10+ million ad auctions per second globally -- each auction deciding in <100ms which ad to show. Deep learning models (DeepFM, DCN-V2, DLRM) process sparse
      categorical features (user ID, ad ID, context) and dense features via embedding layers and feature interaction networks, predicting CTR (click-through rate) with log loss ~0.12-0.15.
    source_title: DeepFM (2017, Huawei) / DCN-V2 (2020, Google) / DLRM (2019, Meta) -- CTR prediction architectures / Google Ad Manager RTB specifications
    source_url: https://arxiv.org/abs/1906.00091
    confidence: high
  - id: af-ai-for-online-advertising-2
    statement: >-
      Google, Meta, and Amazon collectively capture ~60% of the $600B+ global digital advertising market (2025) -- with Meta's ad system processing trillions of predictions daily using GNN-based and
      transformer-based recommendation models that jointly optimize user engagement, advertiser ROI, and platform revenue through multi-objective reinforcement learning.
    source_title: Meta Ads Engineering Blog (2023-2025) / Google Ads AI / Amazon Advertising -- ML systems at trillion-scale / IAB Digital Advertising Revenue Report 2025
    source_url: https://arxiv.org/abs/2008.13535
    confidence: high
primary_sources:
  - id: ps-ai-for-online-advertising-1
    title: "DLRM: Deep Learning Recommendation Model for Personalization and Recommendation Systems"
    type: academic_paper
    year: 2019
    institution: arXiv / Meta AI
    url: https://arxiv.org/abs/1906.00091
  - id: ps-ai-for-online-advertising-2
    title: "DCN-V2: Improved Deep and Cross Network for Web-Scale Learning to Rank Systems"
    type: academic_paper
    year: 2020
    institution: Google Research / KDD
    url: https://arxiv.org/abs/2008.13535
known_gaps:
  - Privacy-preserving advertising after third-party cookie deprecation
  - Causal estimation of ad effectiveness -- did the ad actually cause the conversion?
disputed_statements: []
secondary_sources:
  - title: A Systematic Review and Research Perspective on Recommender Systems
    type: survey_paper
    year: 2022
    authors:
      - multiple
    institution: Journal of Big Data (Springer)
    url: https://doi.org/10.1186/s40537-022-00592-5
  - title: "A Systematic Literature Review on AI-Based Recommendation Systems: Technologies, Applications, and Challenges"
    type: survey_paper
    year: 2024
    authors:
      - multiple
    institution: IEEE Access
    url: https://doi.org/10.1109/ACCESS.2024.3451218
  - title: "Artificial Intelligence in Advertising: Advancements, Challenges, and Ethical Considerations in Targeting, Personalization, Content Creation, and Ad Optimization"
    type: survey_paper
    year: 2023
    authors:
      - multiple
    institution: SAGE Open
    url: https://doi.org/10.1177/21582440231210759
  - title: "Review-based Recommender Systems: A Survey of Approaches, Challenges, and Future Directions"
    type: survey_paper
    year: 2024
    authors:
      - multiple
    institution: arXiv
    url: https://arxiv.org/abs/2405.05562
updated: "2026-05-24"
---
## TL;DR
Online advertising is the financial engine of the internet -- and AI is its brain. Real-time bidding systems make millions of decisions per second about which ad to show, using deep learning to predict click probability, conversion likelihood, and user value. From Google Search Ads to TikTok's feed, AI-powered advertising generates over $600B annually.

## Core Explanation
Online advertising flow: User visits webpage -> ad request sent to exchange (Google Ad Manager, Meta Ads) -> Real-time bidding (RTB) -- advertisers submit bids based on predicted value = p(click) * p(conversion|click) * conversion value -> Ad selection -> Feedback (click or not, converts or not) generates training data. Key ML problem: CTR prediction. Input: user features, ad features, context features. Output: probability of click. Extremely sparse and high-dimensional input (billions of user/ad IDs) requiring massive embedding tables (TB scale).

## Detailed Analysis
CTR architectures: (1) DeepFM -- combine linear (FM) and deep (DNN) components for memorization + generalization; (2) DCN-V2 (Google) -- explicit feature crossing via cross layers computing high-order interactions; (3) DLRM (Meta) -- dense features via MLP + sparse features via embedding lookups + feature interaction via dot product; (4) DIN/DIEN (Alibaba) -- attention-based models weighting user behaviors by relevance to target ad. Calibration: predicted CTR must match empirical CTR (Platt scaling, isotonic regression). Delayed feedback: conversions happen hours/days after clicks. Cold-start: new ads and users have no history -- content-based and few-shot adaptation. Post-cookie era (2024-2026): Google deprecated third-party cookies. Alternatives: first-party data, contextual targeting, Privacy Sandbox APIs (FLEDGE/Topics), and cohort-based advertising.