arXiv subjects

John C. Platt

Publications and source records attributed to John C. Platt.

10 recordsLinked to original sources

arXiv2026

An AI system to help scientists write expert-level empirical software

The cycle of scientific discovery is frequently bottlenecked by the slow, manual creation of software to support computational experiments\cite{hannay2009how}. To address this, we present Empirical Research Assistance (ERA), an AI system that creates expert-level scientific software whose goal is to maximize a quality metric. The system uses a Large Language Model (LLM) and Tree Search (TS)\cite{silver2016mastering} to systematically improve the quality metric and intelligently navigate the large space of possible solutions. ERA achieves expert-level results when it explores and integrates complex research ideas from external sources. The effectiveness of tree search is demonstrated across a diverse range of tasks. In bioinformatics, ERA discovered 40 novel methods for single-cell data analysis that outperformed the top human-developed methods on a public leaderboard. In epidemiology, ERA generated 14 models that outperformed the CDC ensemble and all other individual models for forecasting COVID-19 hospitalizations. ERA also produced expert-level software for geospatial analysis, neural activity prediction in zebrafish, and numerical solution of integrals, and a novel rule-based construction for time series forecasting. By devising and implementing novel solutions to diverse tasks, ERA represents a significant step towards accelerating scientific progress.

cs.AI↗

arXiv2026

Optimized Three-Dimensional Photovoltaic Structures with LLM guided Tree Search

We present a case study for how AI coding systems can be used to generate novel scientific hypotheses. We combine a generic coding agent (Google's AntiGravity) with an LLM-driven tree search algorithm (Empirical Research Assistance / ERA) to autonomously generate high-efficiency three-dimensional photovoltaic (3DPV) structures that overcome losses limiting flat solar panels at mid-latitudes. These structures operate by presenting favorable angles to the sun throughout the day, and for illustrative purposes we focus on optimizing performance for a single solar day. Our workflow begins by using AntiGravity to reproduce calculations \cite{bernardi2012solar} showing that 3DPV can have energy densities much higher than stationary flat PV panels. We use these initial designs as the starting point for large scale tree search, where we seek improved solutions and score them for their diurnal yield. The initial tree search leads to nominally more efficient solutions, yet they are caused by algorithmic reward hacking, arising from non-physical design features such as structurally levitating disconnected tiers and exploitations of the discretizations in the optics solver. To counteract this, we develop a workflow where the coding agent iteratively patches the physics engine with constraints to eliminate reward hacking. With reward-hacking eliminated, ERA discovers a series of designs with various constraints and improved performance, including optimal designs with different fixed collector areas, optimizing zenith tracking and avoiding self shadowing. Combining coding agents with tree search (ERA) provides a powerful platform for scientific discovery, for problems whose solutions can be empirically evaluated with a score function.

cs.CL↗

arXiv2025

Estimating high-resolution albedo for urban applications

Implementation of cool roofs is a high-impact pathway for mitigating heat at both global and city scales. However, while albedo estimates derived from Sentinel-2 are free and globally-available, the 10 m resolution is insufficient to resolve individual roofs. We present methods for increasing the resolution of Sentinel-2 albedo using high-resolution satellite imagery to produce albedo inferences at a 30-cm scale. Validating against high-resolution aerial albedo measurements over Boulder, CO we find improved precision and accuracy relative to Sentinel-2 with an RMSE of 0.04. Applying these methods to 12 global cities, we evaluate the impacts of three cool roof implementation scenarios. We find that cities can see up to a 0.5°C cooling effect from full scale implementation of cool roofs and prioritizing the largest buildings for implementation is a highly effective policy pathway. While Sentinel-2 produces accurate estimates of albedo change at larger scales, high-resolution inferences are required for prioritizing buildings based on their solar radiation management potential. This research demonstrates a scalable implementation of targeted cool roof interventions in neighborhoods with the greatest potential for heat mitigation by enabling actionable, building-level insights.

physics.geo-ph↗

arXiv2023

A scalable system to measure contrail formation on a per-flight basis

Persistent contrails make up a large fraction of aviation's contribution to global warming. We describe a scalable, automated detection and matching (ADM) system to determine from satellite data whether a flight has made a persistent contrail. The ADM system compares flight segments to contrails detected by a computer vision algorithm running on images from the GOES-16 Advanced Baseline Imager. We develop a 'flight matching' algorithm and use it to label each flight segment as a 'match' or 'non-match'. We perform this analysis on 1.6 million flight segments. The result is an analysis of which flights make persistent contrails several orders of magnitude larger than any previous work. We assess the agreement between our labels and available prediction models based on weather forecasts. Shifting air traffic to avoid regions of contrail formation has been proposed as a possible mitigation with the potential for very low cost/ton-CO2e. Our findings suggest that imperfections in these prediction models increase this cost/ton by about an order of magnitude. Contrail avoidance is a cost-effective climate change mitigation even with this factor taken into account, but our results quantify the need for more accurate contrail prediction methods and establish a benchmark for future development.

physics.ao-ph↗

arXiv2022

Efficient approximation of experimental Gaussian boson sampling

Two recent landmark experiments have performed Gaussian boson sampling (GBS) with a non-programmable linear interferometer and threshold detectors on up to 144 output modes (see Refs.~\onlinecite{zhong_quantum_2020,zhong2021phase}). Here we give classical sampling algorithms with better total variation distance and Kullback-Leibler divergence than these experiments and a computational cost quadratic in the number of modes. Our method samples from a distribution that approximates the single-mode and two-mode ideal marginals of the given Gaussian boson sampler, which are calculated efficiently. One implementation sets the parameters of a Boltzmann machine from the calculated marginals using a mean field solution. This is a 2nd order approximation, with the uniform and thermal approximations corresponding to the 0th and 1st order, respectively. The $k$th order approximation reproduces Ursell functions (also known as connected correlations) up to order $k$ with a cost exponential in $k$ and high precision, while the experiment exhibits higher order Ursell functions with lower precision. This methodology, like other polynomial approximations introduced previously, does not apply to random circuit sampling because the $k$th order approximation would simply result in the uniform distribution, in contrast to GBS.

quant-ph↗

arXiv2021

Living-Off-The-Land Command Detection Using Active Learning

In recent years, enterprises have been targeted by advanced adversaries who leverage creative ways to infiltrate their systems and move laterally to gain access to critical data. One increasingly common evasive method is to hide the malicious activity behind a benign program by using tools that are already installed on user computers. These programs are usually part of the operating system distribution or another user-installed binary, therefore this type of attack is called "Living-Off-The-Land". Detecting these attacks is challenging, as adversaries may not create malicious files on the victim computers and anti-virus scans fail to detect them. We propose the design of an Active Learning framework called LOLAL for detecting Living-Off-the-Land attacks that iteratively selects a set of uncertain and anomalous samples for labeling by a human analyst. LOLAL is specifically designed to work well when a limited number of labeled samples are available for training machine learning models to detect attacks. We investigate methods to represent command-line text using word-embedding techniques, and design ensemble boosting classifiers to distinguish malicious and benign samples based on the embedding representation. We leverage a large, anonymized dataset collected by an endpoint security product and demonstrate that our ensemble classifiers achieve an average F1 score of 0.96 at classifying different attack classes. We show that our active learning method consistently improves the classifier performance, as more training data is labeled, and converges in less than 30 iterations when starting with a small number of labeled instances.

cs.CR↗

arXiv2019

Supplementary information for "Quantum supremacy using a programmable superconducting processor"

This is an updated version of supplementary information to accompany "Quantum supremacy using a programmable superconducting processor", an article published in the October 24, 2019 issue of Nature. The main article is freely available at https://www.nature.com/articles/s41586-019-1666-5. Summary of changes since arXiv:1910.11333v1 (submitted 23 Oct 2019): added URL for qFlex source code; added Erratum section; added Figure S41 comparing statistical and total uncertainty for log and linear XEB; new References [1,65]; miscellaneous updates for clarity and style consistency; miscellaneous typographical and formatting corrections.

quant-ph↗

arXiv2019

Tackling Climate Change with Machine Learning

Climate change is one of the greatest challenges facing humanity, and we, as machine learning experts, may wonder how we can help. Here we describe how machine learning can be a powerful tool in reducing greenhouse gas emissions and helping society adapt to a changing climate. From smart grids to disaster management, we identify high impact problems where existing gaps can be filled by machine learning, in collaboration with other fields. Our recommendations encompass exciting research questions as well as promising business opportunities. We call on the machine learning community to join the global effort against climate change.

cs.CY↗

arXiv2015

From Captions to Visual Concepts and Back

This paper presents a novel approach for automatically generating image descriptions: visual detectors, language models, and multimodal similarity models learnt directly from a dataset of image captions. We use multiple instance learning to train visual detectors for words that commonly occur in captions, including many different parts of speech such as nouns, verbs, and adjectives. The word detector outputs serve as conditional inputs to a maximum-entropy language model. The language model learns from a set of over 400,000 image descriptions to capture the statistics of word usage. We capture global semantics by re-ranking caption candidates using sentence-level features and a deep multimodal similarity model. Our system is state-of-the-art on the official Microsoft COCO benchmark, producing a BLEU-4 score of 29.1%. When human judges compare the system captions to ones written by other people on our held-out test set, the system captions have equal or better quality 34% of the time.

cs.CV↗

arXiv2015

Regularized Minimax Conditional Entropy for Crowdsourcing

There is a rapidly increasing interest in crowdsourcing for data labeling. By crowdsourcing, a large number of labels can be often quickly gathered at low cost. However, the labels provided by the crowdsourcing workers are usually not of high quality. In this paper, we propose a minimax conditional entropy principle to infer ground truth from noisy crowdsourced labels. Under this principle, we derive a unique probabilistic labeling model jointly parameterized by worker ability and item difficulty. We also propose an objective measurement principle, and show that our method is the only method which satisfies this objective measurement principle. We validate our method through a variety of real crowdsourcing datasets with binary, multiclass or ordinal labels.

cs.LG↗