ICML: “Test of time” paper shows how times have changed

Amazon scientist’s award-winning paper predates — but later found applications in — the deep-learning revolution.

Amazon researchers have nine new papers at this year’s International Conference on Machine Learning (ICML), one of the top conferences in AI. Matthias Seeger, a principal applied scientist with Amazon Web Services (AWS), is a coauthor on one of them, which reports work led by AWS applied scientist Cuong Nguyen.

But it’s a paper that Seeger cowrote ten years ago that’s one of the conference highlights. On July 1, the ICML awards committee announced that Seeger and his colleagues’ 2010 paper “Gaussian process optimization in the bandit setting: no regret and experimental design” had won the conference’s Test of Time Award, which honors “a paper from ICML ten years ago that has had substantial impact on the field of machine learning, including both research and practice.”

The citation from the award committee begins, “This paper brought together the fields of Bayesian optimization, bandits, and experimental design,” and it “has since cross-fertilized these separate research domains,” Seeger adds.

Matthias Seeger
Matthias Seeger, principal applied scientist

Bayesian-optimization and bandit problems have the same general structure, Seeger explains, but “Bayesian optimization is generally done over continuous input spaces and more complicated functions,” he says. “Multi-armed bandits would normally assume finite spaces and linear or otherwise strongly restricted payoff functions. Maybe because Bayesian optimization is more flexible in this sense, it comes with a lot less solid theory. Multi-armed bandits is a more theoretically grounded area.”

Seeger and his colleagues’ 2010 paper generalized theoretical findings from the multi-armed bandit setting to Bayesian optimization (BO), providing strong performance bounds given particular choices of statistical models. This gave machine learning practitioners greater confidence in techniques they’d arrived at empirically and helped them identify circumstances in which those techniques might be less successful.

In the context of deep learning — which now dominates the field of artificial intelligence — BO is used for hyperparameter tuning, or optimizing structural features of the deep-learning model and parameters of the learning algorithm to maximize the efficacy of training on particular data.

To prove their result for BO, Seeger and his colleagues extended techniques borrowed from a third related field, experimental design. The tools they devised to bridge the related disciplines of BO, multi-armed bandits, and experimental design have proved useful to researchers working in all three; the paper has more than 1,000 citations on Google Scholar, which have helped make Seeger the fourth most highly cited researcher in the field of Bayesian optimization.

Seeger’s coauthors on the 2010 paper are Niranjan Srinivas, now a computational biologist at 10xGenomics; Andreas Krause, now a professor of computer science at ETH Zurich; and Sham Kakade, now a professor in the departments of computer science and statistics at the University of Washington.

With Bayesian optimization, Seeger explains, “you are essentially optimizing a function over some search space without actually knowing what this function looks like. You have to learn about that function as you sample it. But your real goal is finding the function’s maximum, or to sample it nearby.”

“If you sample forever, at some point you will find its optima” he adds. “But since sampling is expensive and takes time, you want to finish as rapidly as possible. So what you are really interested in is to spend as few samples as possible before you converge to something useful, very close to the optimum.”

Temperature samples.png
An example of Seeger and his colleagues’ sample selection procedure, taken from their 2010 paper. The first image (a) represents temperatures in different parts of a building; the two images at right (b and c) represent successive iterations of the procedure, in which individual sensors are briefly activated to take temperature readings (red circles). The black line represents the true temperatures, and the grey areas represent the method’s latest inference of the range of possible temperatures in each region. Crosses indicate points at which readings have already been taken. The procedure selects new sample points with the goal of either maximizing information gain (b) or finding optima (c).

Seeger and his colleagues proved that, under conditions that frequently hold for machine learning problems, the sampling process is guaranteed to converge. But they also showed that the convergence rate depends on specific problem parameters.

In BO, the function that you’re trying to optimize is a random function, Seeger explains. “Every time you plug in a point x, you get a random value f(x),” he says. A standard way to do BO is to model the outputs of the function using a probability distribution. If that distribution is Gaussian — the standard bell curve — then Bayesian optimization is said to use a Gaussian process as a surrogate model.

One of the parameters of the surrogate model is its covariance function, which describes the correlation between changes to function inputs (the x’s) and the resulting changes to the outputs (the f(x)’s). There are several families of covariance function, with an infinite range of functions within each family.

Seeger and his colleagues’ paper quantitatively relates the convergence rate of the function-sampling procedure to the specific choice of covariance function.

“Some choices of covariance function imply smooth functions, which can faithfully be interpolated from measurements nearby,” Seeger says. “Others result in rough functions, for which interpolating even across short distances is an uncertain exercise.”

From theory to practice

Machine learning researchers had conjectured that rougher covariance functions, although they may model reality better, imply slower convergence on an optimum. Seeger and his colleagues’ paper quantified this trade-off precisely. It enabled researchers to decide how much roughness they were willing to sacrifice for faster convergence.

To bound convergence rates, Seeger and his colleagues borrowed techniques from experimental design. “Here you are interested in learning about a function, but not just where its maximum is — learning about it globally, or learning about a model everywhere,” Seeger explains.

A central concept in experimental design is information gain, a quantification of how much information about a function each new sample — each new experiment — confers. Seeger and his colleagues quantified the convergence rate in BO by framing the question in terms of information gain.

“It’s pretty theoretical work,” Seeger says. “It led to a lot of follow-up work because of the links that we could show. There is quite a bit of Gaussian-process theory being brought together here, which partly fuses the different themes together.”

At Amazon, Seeger says, his work on meta-learning and automated machine learning is less theoretical, but his training still stands him in good stead. “I do think that the background that I have from back then is quite useful for me to plan ahead, in a way, and to see whether something looks right or not,” Seeger says. “I think this is quite important, because due to the explosive growth of our field, there are now so many options you can pursue. You really cannot implement all of them and try them out. You need to have some guiding principles.”

Seeger’s transition from more theoretical to more applied work mirrors that of the field itself. In the ten years since he cowrote this award-winning paper, he says, “there’s been a huge change, with the size of the field, the practical applications, the size of the models, the size of the data sets.”

“Back then, maybe because it was smaller, there were ideas being followed that wouldn’t immediately have an application, and you were actually looking into theory quite a bit,” he says. “These days, it’s a lot more driven by empirical applications and empirical work. The field has matured and is very successful in some applications, so obviously you want to focus more on whether an idea is useful in the current context or not. And that’s certainly something that I’ve learned to do a lot more since I joined Amazon.”

Research areas

Related content

US, WA, Seattle
Note that this posting is for a handful of teams within Amazon Robotics. Teams include: Robotics, Computer Vision, Machine Learning, Optimization, and more.Are you excited about building high-performance robotic systems that can perceive and learn to help deliver for customers? The Amazon Robotics team is creating new science products and technologies that make this possible, at Amazon scale. We work at the intersection of computer vision, machine learning, robotic manipulation, navigation, and human-robot interaction.Amazon Robotics is seeking broad, curious applied scientists and engineering interns to join our diverse, full-stack team. In addition to designing, building, and delivering end-to-end robotic systems, our team is responsible for core infrastructure and tools that serve as the backbone of our robotic applications, enabling roboticists, applied scientists, software and hardware engineers to collaborate and deploy systems in the lab and in the field. We will give you the tools and support you need to invent with us in ways that are rewarding, fulfilling and fun. Come join us!A day in the lifeAs an intern you will develop a new algorithm to solve one of the challenging computer vision and manipulation problems in Amazon's robotic warehouses. Your project will fit your academic research experience and interests. You will code and test out your solutions in increasingly realistic scenarios and iterate on the idea with your mentor to find the best solution to the problem.
US, WA, Bellevue
The Global Supply Chain-ACES organization aims to raise the bar on Amazon’s customer experience by delivering holistic solutions for Global Customer Fulfillment that facilitate the effective and efficient movement of product through our supply chain. We develop strategies, processes, material handling and technology solutions, reporting and other mechanisms, which are simple, technology enabled, globally scalable, and locally relevant. We achieve this through cross-functional partnerships, listening to the needs of our customers and prioritizing initiatives to deliver maximum impact across the value chain. Within the organization, our Quality team balances tactical operation with operations partners with global engagement on programs to deliver improved inventory accuracy in our network. The organization is looking for an experienced Principal Data Scientist to partner with senior leadership to develop long term strategic solutions. As a Principal Scientist, they will lead critical initiatives for Global Supply Chain, leveraging complex data analysis and visualization to:a. Collaborate with business teams to define data requirements and processes;b. Automate data pipelines;c. Design, develop, and maintain scalable (automated) reports and dashboards that track progress towards plans;d. Define, track and report program success metrics.e. Serve as a technical science lead on our most demanding, cross-functional projects.
US, MA, North Reading
Are you inspired by invention? Is problem solving through teamwork in your DNA? Do you like the idea of seeing how your work impacts the bigger picture? Answer yes to any of these and you’ll fit right in here at Amazon Robotics. We are a smart team of doers that work passionately to apply cutting edge advances in robotics and software to solve real-world challenges that will transform our customers’ experiences in ways we can’t even imagine yet. We invent new improvements every day. We are Amazon Robotics and we will give you the tools and support you need to invent with us in ways that are rewarding, fulfilling and fun. The Research Science team at Amazon Robotics is seeking interns with a passion for robotic research to work on cutting edge algorithms for robotics. Our team works on challenging and high-impact projects, including allocating resources to complete a million orders a day, coordinating the motion of thousands of robots, autonomous navigation in warehouses, and learning how to grasp all the products Amazon sells. We are seeking internship candidates with backgrounds in computer vision, machine learning, resource allocation, discrete optimization, search, planning/scheduling, and reinforcement learning. As an intern you will develop a new algorithm to solve one of the challenging computer vision and manipulation problems in Amazon's robotic warehouses. Your project will fit your academic research experience and interests. You will code and test out your solutions in increasingly realistic scenarios and iterate on the idea with your mentor to find the best solution to the problem.
US, WA, Seattle
Are you excited about building high-performance robotic systems that can perceive, learn, and act intelligently alongside humans? The Robotics AI team is creating new science products and technologies that make this possible, at Amazon scale. We work at the intersection of computer vision, machine learning, robotic manipulation, navigation, and human-robot interaction.The Amazon Robotics team is seeking broad, curious applied scientists and engineering interns to join our diverse, full-stack team. In addition to designing, building, and delivering end-to-end robotic systems, our team is responsible for core infrastructure and tools that serve as the backbone of our robotic applications, enabling roboticists, applied scientists, software and hardware engineers to collaborate and deploy systems in the lab and in the field. Come join us!
US, WA, Bellevue
Employer: Amazon.com Services LLCPosition: Research Scientist IILocation: Bellevue, WA Multiple Positions Available1. Research, build and implement highly effective and innovative methods in Statistical Modeling, Machine Learning, and other quantitative techniques such as operational research and optimization to deliver algorithms that solve real business problems.2. Take initiative to scope and plan research projects based on roadmap of business owners and enable data-driven solutions. Participate in shaping roadmap for the research team.3. Ensure data quality throughout all stages of acquisition and processing of the data, including such areas as data sourcing/collection, ground truth generation, data analysis, experiment, evaluation and visualization etc.4. Navigate a variety of data sources, understand the business reality behind large-scale data and develop meaningful science solutions.5. Partner closely with product or/and program owners, as well as scientists and engineers in cross-functional teams with a clear path to business impact and deliver on demanding projects.6. Present proposals and results in a clear manner backed by data and coupled with conclusions to business customers and leadership team with various levels of technical knowledge, educating them about underlying systems, as well as sharing insights.7. Perform experiments to validate the feature additions as requested by domain expert teams.8. Some telecommuting benefits available.The pay range for this position in Bellevue, WA is $136,000-$184,000 (yr); however, base pay offered may vary depending on job-related knowledge, skills, and experience. A sign-on bonus and restricted stock units may be provided as part of the compensation package, in addition to a full range of medical, financial, and/or other benefits, dependent on the position offered. This information is provided by the Washington Equal Pay Act. Base pay information is based on market location. Applicants should apply via Amazon's internal or external careers site.#0000
US, VA, Arlington
The Central Science Team within Amazon’s People Experience and Technology org (PXTCS) uses economics, behavioral science, statistics, and machine learning to proactively identify mechanisms and process improvements which simultaneously improve Amazon and the lives, well-being, and the value of work to Amazonians. We are an interdisciplinary team, which combines the talents of science and engineering to develop and deliver solutions that measurably achieve this goal. As Director for PXT Central Science Technology, you will be responsible for leading multiple teams through rapidly evolving complex demands and define, develop, deliver and execute on our science roadmap and vision. You will provide thought leadership to scientists and engineers to invent and implement scalable machine learning recommendations and data driven algorithms supporting flexible UI frameworks. You will manage and be responsible for delivering some of our most strategic technical initiatives. You will design, develop and operate new, highly scalable software systems that support Amazon’s efforts to be Earth’s Best Employer and have a significant impact on Amazon’s commitment to our employees and communities where we both serve and employ 1.3 million Amazonians. As Director of Applied Science, you will be part of the larger technical leadership community at Amazon. This community forms the backbone of the company, plays a critical role in the broad business planning, works closely with senior executives to develop business targets and resource requirements, influences our long-term technical and business strategy, helps hire and develop engineering leaders and developers, and ultimately enables us to deliver engineering innovations.This role is posted for Arlington, VA, but we are flexible on location at many of our offices in the US and Canada.
US, VA, Arlington
Employer: Amazon.com Services LLCPosition: Data Scientist IILocation: Arlington, VAMultiple Positions Available1. Manage and execute entire projects or components of large projects from start to finish including data gathering and manipulation, synthesis and modeling, problem solving, and communication of insights and recommendations.2. Oversee the development and implementation of data integration and analytic strategies to support population health initiatives.3. Leverage big data to explore and introduce areas of analytics and technologies.4. Analyze data to identify opportunities to impact populations.5. Perform advanced integrated comprehensive reporting, consultative, and analytical expertise to provide healthcare cost and utilization data and translate findings into actionable information for internal and external stakeholders.6. Oversee the collection of data, ensuring timelines are met, data is accurate and within established format.7. Act as a data and technical resource and escalation point for data issues, ensuring they are brought to resolution.8. Serve as the subject matter expert on health care benefits data modeling, system architecture, data governance, and business intelligence tools. #0000
US, TX, Dallas
Employer: Amazon.com Services LLCPosition: Data Scientist II (multiple positions available)Location: Dallas, TX Multiple Positions Available:1. Assist customers to deliver Machine Learning (ML) and Deep Learning (DL) projects from beginning to end, by aggregating data, exploring data, building and validating predictive models, and deploying completed models to deliver business impact to the organization;2. Apply understanding of the customer’s business need and guide them to a solution using AWS AI Services, AWS AI Platforms, AWS AI Frameworks, and AWS AI EC2 Instances;3. Use Deep Learning frameworks like MXNet, PyTorch, Caffe 2, Tensorflow, Theano, CNTK, and Keras to help our customers build DL models;4. Research, design, implement and evaluate novel computer vision algorithms and ML/DL algorithms;5. Work with data architects and engineers to analyze, extract, normalize, and label relevant data;6. Work with DevOps engineers to help customers operationalize models after they are built;7. Assist customers with identifying model drift and retraining models;8. Research and implement novel ML and DL approaches, including using FPGA;9. Develop computer vision and machine learning methods and algorithms to address real-world customer use-cases; and10. Design and run experiments, research new algorithms, and work closely with engineers to put algorithms and models into practice to help solve customers' most challenging problems.11. Approximately 15% domestic and international travel required.12. Telecommuting benefits are available.#0000
US, WA, Seattle
MULTIPLE POSITIONS AVAILABLECompany: AMAZON.COM SERVICES LLCPosition Title: Manager III, Data ScienceLocation: Bellevue, WashingtonPosition Responsibilities:Manage a team of data scientists working to build large-scale, technical solutions to increase effectiveness of Amazon Fulfillment systems. Define key business goals and map them to the success of technical solutions. Aggregate, analyze and model data from multiple sources to inform business decisions. Manage and quantify improvement in the customer experience resulting from research outcomes. Develop and manage a long-term research vision and portfolio of research initiatives, with algorithms and models that to be integrated in production systems. Hire and mentor junior scientists.Amazon.com is an Equal Opportunity-Affirmative Action Employer – Minority / Female / Disability / Veteran / Gender Identity / Sexual Orientation #0000
US, VA, Arlington
MULTIPLE POSITIONS AVAILABLECompany: AMAZON.COM SERVICES LLCPosition Title: Data Scientist IILocation: Arlington, VirginiaPosition Responsibilities:Design and implement scalable and reliable approaches to support or automate decision making throughout the business. Apply a range of data science techniques and tools combined with subject matter expertise to solve difficult business problems and cases in which the solution approach is unclear. Acquire data by building the necessary SQL / ETL queries. Import processes through various company specific interfaces for accessing Oracle, RedShift, and Spark storage systems. Build relationships with stakeholders and counterparts. Analyze data for trends and input validity by inspecting univariate distributions, exploring bivariate relationships, constructing appropriate transformations, and tracking down the source and meaning of anomalies. Build models using statistical modeling, mathematical modeling, econometric modeling, network modeling, social network modeling, natural language processing, machine learning algorithms, genetic algorithms, and neural networks. Validate models against alternative approaches, expected and observed outcome, and other business defined key performance indicators. Implement models that comply with evaluations of the computational demands, accuracy, and reliability of the relevant ETL processes at various stages of production.Amazon.com is an Equal Opportunity-Affirmative Action Employer – Minority / Female / Disability / Veteran / Gender Identity / Sexual Orientation #0000