arXiv subjects

Jeffrey Wong

Publications and source records attributed to Jeffrey Wong.

12 recordsLinked to original sources

arXiv2026

Closing the Loop: A Software Framework for AI to Support Business Decision Making

Create an idea, prototype it, evaluate if users like it, then learn. It is the circle of business. If AI can operate in all parts of the circle, it will enable rapid iteration and learning speeds for businesses. Experiment platforms that deploy experiments to evaluate return on investment for businesses are abundant, but systems that help businesses learn personalization, mechanisms, and what to ideate next, are rare. Among technologies that do exist, they cannot be well orchestrated in a single software interface that can be safely and efficiently leveraged by an AI agent. These challenges make it difficult to teach an AI agent how to learn within a robust experimentation framework, and difficult for an AI agent to operate and iterate for the business. We offer a two part solution: one half that is rooted in mathematical reductions to contain complexity, and one half that is rooted in software design to optimize for orchestration, software safety, and multiplicity. Our solution, a software framework, moves beyond the simple treatment effect computed as a difference in means. To create a better understanding of a business and its customers, we enrich causal analysis with heterogeneous effects, policy algorithms, mediation analysis, and forecasts of effects. To have an AI complete the iteration cycle faster, we further enrich the analysis with variance reduction and anytime valid inference. The enrichments are made compatible across different types of experiments, and are presented in a single software interface that is usable in an AI agent. We evaluate the approach on various objectives in experiment analysis, and show that the framework improves code correctness, reduces lines of code, and is more performant than a baseline analysis constructed by a vanilla agent.

cs.SE↗

arXiv2025

ARIES: Autonomous Reasoning with LLMs on Interactive Thought Graph Environments

Recent research has shown that LLM performance on reasoning tasks can be enhanced by scaling test-time compute. One promising approach, particularly with decomposable problems, involves arranging intermediate solutions as a graph on which transformations are performed to explore the solution space. However, prior works rely on pre-determined, task-specific transformation schedules which are subject to a set of searched hyperparameters. In this work, we view thought graph transformations as actions in a Markov decision process, and implement policy agents to drive effective action policies for the underlying reasoning LLM agent. In particular, we investigate the ability for another LLM to act as a policy agent on thought graph environments and introduce ARIES, a multi-agent architecture for reasoning with LLMs. In ARIES, reasoning LLM agents solve decomposed subproblems, while policy LLM agents maintain visibility of the thought graph states, and dynamically adapt the problem-solving strategy. Through extensive experiments, we observe that using off-the-shelf LLMs as policy agents with no supervised fine-tuning (SFT) can yield up to $29\%$ higher accuracy on HumanEval relative to static transformation schedules, as well as reducing inference costs by $35\%$ and avoid any search requirements. We also conduct a thorough analysis of observed failure modes, highlighting that limitations on LLM sizes and the depth of problem decomposition can be seen as challenges to scaling LLM-guided reasoning.

cs.AI↗

arXiv2024

Delta Vectors Unify the Computation for Linear Model Treatment Effects

The science of cause and effect is extremely sophisticated and extremely hard to scale. Using a controlled experiment, scientists get rich insights by analyzing global effects, effects in different segments, and trends in effects over time. They use propensity scores to project external validity. To support the analysis of relative effects, scientists derive challenging ratio distributions. While the analytical capabilities in experimentation are advancing, we require new innovation within engineering and computational causal inference to enable an experimentation platform to make analyses performant and scalable. Of significant importance: we must unify the computing strategy for these models so that they can be consistently applied across experiments. In doing so, the industry can make significant progress towards developing a flywheel that unifies and accelerates the evaluation and roll out of experiments. In order to support unified computation, this paper introduces baseline vectors and delta vectors as common structure for estimating treatment effects. This common structure allows many statistics to be subsumed into a single API. The nature of its algebraic formulation allows linear algebra libraries to vectorize and optimize its performance, creating a single and efficient tool to support the many innovations in experimentation.

stat.CO↗

arXiv2024

Understanding Public Safety Trends in Calgary through data mining

This paper utilizes statistical data from various open datasets in Calgary to to uncover patterns and insights for community crimes, disorders, and traffic incidents. Community attributes like demographics, housing, and pet registration were collected and analyzed through geospatial visualization and correlation analysis. Strongly correlated features were identified using the chi-square test, and predictive models were built using association rule mining and machine learning algorithms. The findings suggest that crime rates are closely linked to factors such as population density, while pet registration has a smaller impact. This study offers valuable insights for city managers to enhance community safety strategies.

cs.CY↗

arXiv2022

Incrementality Bidding and Attribution

The causal effect of showing an ad to a potential customer versus not, commonly referred to as "incrementality", is the fundamental question of advertising effectiveness. In digital advertising three major puzzle pieces are central to rigorously quantifying advertising incrementality: ad buying/bidding/pricing, attribution, and experimentation. Building on the foundations of machine learning and causal econometrics, we propose a methodology that unifies these three concepts into a computationally viable model of both bidding and attribution which spans the randomization, training, cross validation, scoring, and conversion attribution of advertising's causal effects. Implementation of this approach is likely to secure a significant improvement in the return on investment of advertising.

cs.LG↗

arXiv2022

Addressing Hidden Imperfections in Online Experimentation

Technology companies are increasingly using randomized controlled trials (RCTs) as part of their development process. Despite having fine control over engineering systems and data instrumentation, these RCTs can still be imperfectly executed. In fact, online experimentation suffers from many of the same biases seen in biomedical RCTs including opt-in and user activity bias, selection bias, non-compliance with the treatment, and more generally, challenges in the ability to test the question of interest. The result of these imperfections can lead to a bias in the estimated causal effect, a loss in statistical power, an attenuation of the effect, or even a need to reframe the question that can be answered. This paper aims to make practitioners of experimentation more aware of imperfections in technology-industry RCTs, which can be hidden throughout the engineering stack or in the design process.

cs.SE↗

arXiv2021

You Only Compress Once: Optimal Data Compression for Estimating Linear Models

Linear models are used in online decision making, such as in machine learning, policy algorithms, and experimentation platforms. Many engineering systems that use linear models achieve computational efficiency through distributed systems and expert configuration. While there are strengths to this approach, it is still difficult to have an environment that enables researchers to interactively iterate and explore data and models, as well as leverage analytics solutions from the open source community. Consequently, innovation can be blocked. Conditionally sufficient statistics is a unified data compression and estimation strategy that is useful for the model development process, as well as the engineering deployment process. The strategy estimates linear models from compressed data without loss on the estimated parameters and their covariances, even when errors are autocorrelated within clusters of observations. Additionally, the compression preserves almost all interactions with the the original data, unlocking better productivity for both researchers and engineering systems.

cs.LG↗

arXiv2020

Success Stories from a Democratized Experimentation Platform

We demonstrate the effectiveness of democratization and efficient computation as key concepts of our experimentation platform (XP) by presenting four new models supported by the platform: 1) Weighted least squares, 2) Quantile bootstrapping, 3) Bayesian shrinkage, and 4) Dynamic treatment effects. Each model is motivated by a specific business problem but is generalizable and extensible. The modular structure of our platform allows independent innovation on statistical and computational methods. In practice, a technical symbiosis is created where increasingly advanced user contributions inspire innovations to the software that in turn enable further methodological improvements. This cycle adds further value to how the XP contributes to business solutions.

stat.ME↗

arXiv2019

Efficient Computation for Centered Linear Regression with Sparse Inputs

Regression with sparse inputs is a common theme for large scale models. Optimizing the underlying linear algebra for sparse inputs allows such models to be estimated faster. At the same time, centering the inputs has benefits in improving the interpretation and convergence of the model. However, centering the data naturally makes sparse data become dense, limiting opportunities for optimization. We propose an efficient strategy that estimates centered regression while taking advantage of sparse structure in data, improving computational performance and decreasing the memory footprint of the estimator.

stat.CO↗

arXiv2019

Efficient Computation of Linear Model Treatment Effects in an Experimentation Platform

Linear models are a core component for statistical software that analyzes treatment effects. They are used in experimentation platforms where analysis is automated, as well as scientific studies where analysis is done locally and manually. However, the implementations of linear models for online platforms and for local analysis are often different due to differing demands in scalability. To improve leverage, we developed one library that powers both modes of analysis, consolidates many forms of treatment effects from linear models, and optimizes the computational performance of the software. The highly performant algorithms allow the library to run linear models at Netflix's scale and reply to live queries for treatment effects from an online service without using distributed systems in seconds. Wrapping these algorithms in a library that can be used interactively make large scale linear models accessible for both online platforms and scientists to use locally.

stat.CO↗

arXiv2019

Engineering for a Science-Centric Experimentation Platform

Netflix is an internet entertainment service that routinely employs experimentation to guide strategy around product innovations. As Netflix grew, it had the opportunity to explore increasingly specialized improvements to its service, which generated demand for deeper analyses supported by richer metrics and powered by more diverse statistical methodologies. To facilitate this, and more fully harness the skill sets of both engineering and data science, Netflix engineers created a science-centric experimentation platform that leverages the expertise of data scientists from a wide range of backgrounds by allowing them to make direct code contributions in the languages used by scientists (Python and R). Moreover, the same code that runs in production is able to be run locally, making it straightforward to explore and graduate both metrics and causal inference methodologies directly into production services. In this paper, we utilize a case-study research method to provide two main contributions. Firstly, we report on the architecture of this platform, with a special emphasis on its novel aspects: how it supports science-centric end-to-end workflows without compromising engineering requirements. Secondly, we describe its approach to causal inference, which leverages the potential outcomes conceptual framework to provide a unified abstraction layer for arbitrary statistical models and methodologies.

cs.SE↗

arXiv2018

Spatial mapping and analysis of aerosols during a forest fire using computational mobile microscopy

Forest fires are a major source of particulate matter (PM) air pollution on a global scale. The composition and impact of PM are typically studied using only laboratory instruments and extrapolated to real fire events owing to a lack of analytical techniques suitable for field-settings. To address this and similar field test challenges, we developed a mobile-microscopy and machine-learning-based air quality monitoring platform called c-Air, which can perform air sampling and microscopic analysis of aerosols in an integrated portable device. We tested its performance for PM sizing and morphological analysis during a recent forest fire event in La Tuna Canyon Park by spatially mapping the PM. The result shows that with decreasing distance to the fire site, the PM concentration increases dramatically, especially for particles smaller than 2 microns. Image analysis from the c-Air portable device also shows that the increased PM is comparatively strongly absorbing and asymmetric, with an aspect ratio of 0.5-0.7. These PM features indicate that a major portion of the PM may be open-flame-combustion-generated element carbon soot-type particles. This initial small-scale experiment shows that c-Air has some potential for forest fire monitoring.

physics.ins-det↗