Deep & Cross Network for Ad Click Predictions
Author:
Ruoxi Wang, Institute for Computational and Mathematical Engineering, Stanford University
Abstract:
Feature engineering has been the key to the success of many prediction models. However, the process is nontrivial and often requires manual feature engineering or exhaustive searching. DNNs are able to automatically learn feature interactions; however, they generate all the interactions implicitly, and are not necessarily efficient in learning all types of cross features. In this paper, we propose the Deep & Cross Network (DCN) which keeps the benefits of a DNN model, and beyond that, it introduces a novel cross network that is more efficient in learning certain bounded-degree feature interactions. In particular, DCN explicitly applies feature crossing at each layer, requires no manual feature engineering, and adds negligible extra complexity to the DNN model. Our experimental results have demonstrated its superiority over the state-of-art algorithms on the CTR prediction dataset and dense classification dataset, in terms of both model accuracy and memory usage.
More on http://www.kdd.org/kdd2017/
KDD2017 Conference is published on http://videolectures.net/
Видео Deep & Cross Network for Ad Click Predictions канала KDD2017 video
Ruoxi Wang, Institute for Computational and Mathematical Engineering, Stanford University
Abstract:
Feature engineering has been the key to the success of many prediction models. However, the process is nontrivial and often requires manual feature engineering or exhaustive searching. DNNs are able to automatically learn feature interactions; however, they generate all the interactions implicitly, and are not necessarily efficient in learning all types of cross features. In this paper, we propose the Deep & Cross Network (DCN) which keeps the benefits of a DNN model, and beyond that, it introduces a novel cross network that is more efficient in learning certain bounded-degree feature interactions. In particular, DCN explicitly applies feature crossing at each layer, requires no manual feature engineering, and adds negligible extra complexity to the DNN model. Our experimental results have demonstrated its superiority over the state-of-art algorithms on the CTR prediction dataset and dense classification dataset, in terms of both model accuracy and memory usage.
More on http://www.kdd.org/kdd2017/
KDD2017 Conference is published on http://videolectures.net/
Видео Deep & Cross Network for Ad Click Predictions канала KDD2017 video
Показать
Комментарии отсутствуют
Информация о видео
Другие видео канала
DCN V2:Improved Deep &Cross Network and Practical Lessons for Web-scale Learning to Rank SystemsLearning to Make StuffStacked Ensemble Models and Data Science CompetitionsChris Evans on Real-Time Click Prediction @ AdRollProfit Maximization for Online Advertising Demand-Side PlatformRecommender Systems and Deep Models at EtsyA/B Testing in Networks with Adversarial MembersLesson 2.0: Understanding CTRDeep Learning In 5 Minutes | What Is Deep Learning? | Deep Learning Explained Simply | SimplilearnCultural CreativityAirbnb System Design | Booking.com System Design | System Design Interview QuestionDeep & Cross Network (Building recommendation systems with TensorFlow)New432Hz The DEEPEST Healing Let Go Of All Negative Energy Healing MeditationIntro to Feature Engineering with TensorFlow - Machine Learning Recipes #9432 Hz - Deep Healing Music for The Body & Soul - DNA Repair, Relaxation Music, Meditation MusicWide & Deep Learning with TensorFlow - Machine LearningMentoring Session Panel直观讲解因子分解机Factorization MachineCustomer churn prediction using ANN | Deep Learning Tutorial 18 (Tensorflow2.0, Keras & Python)CS224W: Machine Learning with Graphs | 2021 | Lecture 12.1-Fast Neural Subgraph Matching & Counting