Research Scientist Intern, Ads Quality (PhD)

Meta was built to help people connect and share, and over the last decade our tools have played a critical part in changing how people around the world communicate with one another. With over a billion people using the service and more than fifty offices around the globe, a career at Meta offers countless ways to make an impact in a fast growing organization.

Meta is seeking Research Interns to join the Ads Quality team within Facebook Ads Recommendations. Ads Quality is the "user value" team in Facebook Monetization: our mission is to deliver long-term value for both people and businesses by showing ads people love. Every time someone opens or scrolls Facebook, we help decide which ads to show, how many, and where, so that ads create value for advertisers without eroding the experience that brings people back. Our interns frame practical, high-impact problems in ads recommendation as rigorous machine learning problems, and bring their own expertise to build new solutions with guidance from our research scientists and software engineers.

More concretely, today's ranking systems score each ad on its immediate outcome, such as a click or a conversion. But every impression and engagement also changes what a person does next: a relevant ad can build interest and make future ads more valuable, while a poor one can shorten a session or push people toward ignoring ads altogether. Learning this long-term value is hard. Long-horizon outcomes are sparse and delayed, the value "without the event" is never observed, user actions cannot be randomized, and ads and organic content affect each other. If you are excited about combining causal inference, sequence modeling, and reinforcement learning to teach a recommender system the long-term consequences of its decisions, at the scale of billions of people, come join us!

Our internships are twelve (12) to sixteen (16), or twenty-four (24) weeks long and we have various start dates throughout the year.

Responsibilities

Perform research to advance the science and technology of long-term value modeling in large-scale recommendation systems.
Develop novel causal and sequential machine learning methods (e.g.reward design and reinforcement learning models) that estimate the long-term effect of ads on people's future engagement and value.
Design offline evaluations and online experiments that test whether learned long-term value signals improve outcomes for people and advertisers.
Analyze and improve the efficiency, scalability, and stability of deployed ranking systems.
Collaborate with researchers and cross-functional partners, including communicating research plans, progress, and results.
Publish research results and contribute to research that can be applied to Meta product development.

Qualifications

Currently has or is in the process of obtaining a Ph.D. degree in Computer Science, Machine Learning, Statistics, Operations Research, Economics, or a relevant technical field
Must obtain work authorization in country of employment at the time of hire and maintain ongoing work authorization during employment
Experience with Python, C++, C, Java or other related language
Experience with deep learning frameworks such as PyTorch or TensorFlow
Experience building systems based on machine learning or deep learning methods
Research experience in one or more of: causal inference, recommender systems, reinforcement learning, or sequence modeling Intent to return to the degree program after the completion of the internship/co-op
Proven track record of solid research achievements as demonstrated by grants, fellowships, patents, as well as publications at leading conferences such as NeurIPS, ICML, ICLR, KDD, RecSys, WWW, and WSDM
Research or project experience in reinforcement learning, such as offline RL, contextual bandits, off-policy evaluation, or value-function and temporal-difference learning, and its application to recommendation or advertising
Experience with causal inference for decision-making, such as uplift modeling, heterogeneous treatment effects, doubly robust estimation, instrumental variables, or regression discontinuity
Experience with large-scale sequence models or generative recommenders (e.g., Transformer-based user sequence models)
Demonstrated software development experience via tech internships, work experience, coding competitions, or widely used contributions in open source machine-learning repositories
Experience working and communicating cross-functionally in a fast-paced team environment. Ideal candidates should have the ability to quickly understand and identify the research opportunities behind real-world applications, select the appropriate ML methods to explore, and proactively drive the iterations based on clear analysis of the current results