Mountain View, California, United States
Contact Info
8K followers
500+ connections
Articles by Joaquin
-
Our Responsible AI Principles in Practice
Our Responsible AI Principles in Practice
By Joaquin Quiñonero Candela
Activity
-
After an incredible journey spanning a little over two years, the time has come for me to bid farewell to NVIDIA. It has been an honor and privilege…
After an incredible journey spanning a little over two years, the time has come for me to bid farewell to NVIDIA. It has been an honor and privilege…
Liked by Joaquin Quiñonero Candela
-
Here's another great opportunity at my team at Google. We are looking for a Product Director to lead AI Frameworks at Google including JAX and Keras.…
Here's another great opportunity at my team at Google. We are looking for a Product Director to lead AI Frameworks at Google including JAX and Keras.…
Liked by Joaquin Quiñonero Candela
-
US AISI announces it's all-star leadership - the new organization, housed within National Institute of Standards and Technology (NIST) has announced…
US AISI announces it's all-star leadership - the new organization, housed within National Institute of Standards and Technology (NIST) has announced…
Liked by Joaquin Quiñonero Candela
Publications
-
Practical Lessons from Predicting Clicks on Ads at Facebook
ADKDD
Online advertising allows advertisers to only bid and pay for measurable user responses, such as clicks on ads. As a consequence, click prediction systems are central to most online advertising systems. With over 750 million daily active users and over 1 million active advertisers, predicting clicks on Facebook ads is a challenging machine learning task. In this paper we introduce a model which combines decision trees with logistic regression, outperforming either of these methods on its own by…
Online advertising allows advertisers to only bid and pay for measurable user responses, such as clicks on ads. As a consequence, click prediction systems are central to most online advertising systems. With over 750 million daily active users and over 1 million active advertisers, predicting clicks on Facebook ads is a challenging machine learning task. In this paper we introduce a model which combines decision trees with logistic regression, outperforming either of these methods on its own by over 3%, an improvement with significant impact to the overall system performance. We then explore how a number of fundamental parameters impact the final prediction performance of our system. Not surprisingly, the most important thing is to have the right features: those capturing historical information about the user or ad dominate other types of features. Once we have the right features and the right model (decisions trees plus logistic regression), other factors play small roles (though even small improvements are important at scale). Picking the optimal handling for data freshness, learning rate schema and data sampling improve the model slightly, though much less than adding a high-value feature, or picking the right model to begin with.
Other authorsSee publication -
Practical Lessons from Predicting Clicks on Ads at Facebook
ADKDD
Online advertising allows advertisers to only bid and pay for measurable user responses, such as clicks on ads. As a consequence, click prediction systems are central to most online advertising systems. With over 750 million daily active users and over 1 million active advertisers, predicting clicks on Facebook ads is a challenging machine learning task. In this paper we introduce a model which combines decision trees with logistic regression, outperforming either of these methods on its own by…
Online advertising allows advertisers to only bid and pay for measurable user responses, such as clicks on ads. As a consequence, click prediction systems are central to most online advertising systems. With over 750 million daily active users and over 1 million active advertisers, predicting clicks on Facebook ads is a challenging machine learning task. In this paper we introduce a model which combines decision trees with logistic regression, outperforming either of these methods on its own by over 3%, an improvement with significant impact to the overall system performance. We then explore how a number of fundamental parameters impact the final prediction performance of our system. Not surprisingly, the most important thing is to have the right features: those capturing historical information about the user or ad dominate other types of features. Once we have the right features and the right model (decisions trees plus logistic regression), other factors play small roles (though even small improvements are important at scale). Picking the optimal handling for data freshness, learning rate schema and data sampling improve the model slightly, though much less than adding a high-value feature, or picking the right model to begin with.
Other authorsSee publication
Projects
-
The Path of Go
-
An Xbox Live Arcade game based on ‘Go’, the ancient Chinese board game. The game was developed using pioneering artificial intelligence software at Microsoft Research Cambridge.
Other creatorsSee project -
The Path of Go
-
An Xbox Live Arcade game based on ‘Go’, the ancient Chinese board game. The game was developed using pioneering artificial intelligence software at Microsoft Research Cambridge.
Other creatorsSee project
Languages
-
French
Native or bilingual proficiency
-
Spanish
Native or bilingual proficiency
-
German
Professional working proficiency
Recommendations received
2 people have recommended Joaquin
Join now to viewPeople also viewed
Explore collaborative articles
We’re unlocking community knowledge in a new way. Experts add insights directly into each article, started with the help of AI.
Explore More