arXiv subjects

Matteo Chinazzi

Publications and source records attributed to Matteo Chinazzi.

13 recordsLinked to original sources

arXiv2025

SimulRAG: Simulator-based RAG for Grounding LLMs in Long-form Scientific QA

Large Language Models (LLMs) show promise in generating long-form scientific explanations that synthesize evidence and connect multiple factors. However, in long-form scientific question answering, LLMs often hallucinate, producing unsupported or inconsistent claims. Retrieval-Augmented Generation (RAG) improves trustworthiness by grounding generation in external sources; scientific simulators are valuable because they can validate quantitative hypotheses and capture evolving dynamics. Yet simulation-based RAG is non-trivial due to two challenges: how to retrieve from scientific simulators, and how to efficiently verify and update long-form answers. To overcome these challenges, we propose SimulRAG, a simulator-based RAG framework with a generalized retrieval interface that translates between text and simulator parameters/outputs. SimulRAG further introduces claim-level generation with uncertainty estimation and simulator boundary assessment (UE+SBA) to selectively verify and update claims. Unlike tool-first or holistic answer revision, it first elicits diverse answers without retrieval and then grounds uncertain, simulator-verifiable atomic claims with simulator evidence. We also release a long-form scientific QA benchmark spanning climate science, epidemiology, and urban planning, with ground truth verified by simulations and human annotators. Experiments show SimulRAG improves informativeness by 30.4% and factuality by 16.3% over the strongest adapted RAG baselines, while UE+SBA enhances claim-level efficiency and quality.

cs.CL↗

arXiv2023

Disentangled Multi-Fidelity Deep Bayesian Active Learning

To balance quality and cost, various domain areas of science and engineering run simulations at multiple levels of sophistication. Multi-fidelity active learning aims to learn a direct mapping from input parameters to simulation outputs at the highest fidelity by actively acquiring data from multiple fidelity levels. However, existing approaches based on Gaussian processes are hardly scalable to high-dimensional data. Deep learning-based methods often impose a hierarchical structure in hidden representations, which only supports passing information from low-fidelity to high-fidelity. These approaches can lead to the undesirable propagation of errors from low-fidelity representations to high-fidelity ones. We propose a novel framework called Disentangled Multi-fidelity Deep Bayesian Active Learning (D-MFDAL), which learns the surrogate models conditioned on the distribution of functions at multiple fidelities. On benchmark tasks of learning deep surrogates of partial differential equations including heat equation, Poisson's equation and fluid simulations, our approach significantly outperforms state-of-the-art in prediction accuracy and sample efficiency.

cs.LG↗

arXiv2022

Characterizing collective physical distancing in the U.S. during the first nine months of the COVID-19 pandemic

The COVID-19 pandemic offers an unprecedented natural experiment providing insights into the emergence of collective behavioral changes of both exogenous (government mandated) and endogenous (spontaneous reaction to infection risks) origin. Here, we characterize collective physical distancing -- mobility reductions, minimization of contacts, shortening of contact duration -- in response to the COVID-19 pandemic in the pre-vaccine era by analyzing de-identified, privacy-preserving location data for a panel of over 5.5 million anonymized, opted-in U.S. devices. We define five indicators of users' mobility and proximity to investigate how the emerging collective behavior deviates from the typical pre-pandemic patterns during the first nine months of the COVID-19 pandemic. We analyze both the dramatic changes due to the government mandated mitigation policies and the more spontaneous societal adaptation into a new (physically distanced) normal in the fall 2020. The indicators defined here allow the quantification of behavior changes across the rural/urban divide and highlight the statistical association of mobility and proximity indicators with metrics characterizing the pandemic's social and public health impact such as unemployment and deaths. This study provides a framework to study massive social distancing phenomena with potential uses in analyzing and monitoring the effects of pandemic mitigation plans at the national and international level.

physics.soc-ph↗

arXiv2022

The unequal effects of the health-economy tradeoff during the COVID-19 pandemic

The potential tradeoff between health outcomes and economic impact has been a major challenge in the policy making process during the COVID-19 pandemic. Epidemic-economic models designed to address this issue are either too aggregate to consider heterogeneous outcomes across socio-economic groups, or, when sufficiently fine-grained, not well grounded by empirical data. To fill this gap, we introduce a data-driven, granular, agent-based model that simulates epidemic and economic outcomes across industries, occupations, and income levels with geographic realism. The key mechanism coupling the epidemic and economic modules is the reduction in consumption demand due to fear of infection. We calibrate the model to the first wave of COVID-19 in the New York metropolitan area, showing that it reproduces key epidemic and economic statistics, and then examine counterfactual scenarios. We find that: (a) both high fear of infection and strict restrictions similarly harm the economy but reduce infections; (b) low-income workers bear the brunt of both the economic and epidemic harm; (c) closing non-customer-facing industries such as manufacturing and construction only marginally reduces the death toll while considerably increasing unemployment; and (d) delaying the start of protective measures does little to help the economy and worsens epidemic outcomes in all scenarios. We anticipate that our model will help designing effective and equitable non-pharmaceutical interventions that minimize disruptions in the face of a novel pandemic.

econ.GN↗

arXiv2022

Multi-fidelity Hierarchical Neural Processes

Science and engineering fields use computer simulation extensively. These simulations are often run at multiple levels of sophistication to balance accuracy and efficiency. Multi-fidelity surrogate modeling reduces the computational cost by fusing different simulation outputs. Cheap data generated from low-fidelity simulators can be combined with limited high-quality data generated by an expensive high-fidelity simulator. Existing methods based on Gaussian processes rely on strong assumptions of the kernel functions and can hardly scale to high-dimensional settings. We propose Multi-fidelity Hierarchical Neural Processes (MF-HNP), a unified neural latent variable model for multi-fidelity surrogate modeling. MF-HNP inherits the flexibility and scalability of Neural Processes. The latent variables transform the correlations among different fidelity levels from observations to latent space. The predictions across fidelities are conditionally independent given the latent states. It helps alleviate the error propagation issue in existing methods. MF-HNP is flexible enough to handle non-nested high dimensional data at different fidelity levels with varying input and output dimensions. We evaluate MF-HNP on epidemiology and climate modeling tasks, achieving competitive performance in terms of accuracy and uncertainty estimation. In contrast to deep Gaussian Processes with only low-dimensional (< 10) tasks, our method shows great promise for speeding up high-dimensional complex simulations (over 7000 for epidemiology modeling and 45000 for climate modeling).

cs.LG↗

arXiv2021

Quantifying Uncertainty in Deep Spatiotemporal Forecasting

Deep learning is gaining increasing popularity for spatiotemporal forecasting. However, prior works have mostly focused on point estimates without quantifying the uncertainty of the predictions. In high stakes domains, being able to generate probabilistic forecasts with confidence intervals is critical to risk assessment and decision making. Hence, a systematic study of uncertainty quantification (UQ) methods for spatiotemporal forecasting is missing in the community. In this paper, we describe two types of spatiotemporal forecasting problems: regular grid-based and graph-based. Then we analyze UQ methods from both the Bayesian and the frequentist point of view, casting in a unified framework via statistical decision theory. Through extensive experiments on real-world road network traffic, epidemics, and air quality forecasting tasks, we reveal the statistical and computational trade-offs for different UQ methods: Bayesian methods are typically more robust in mean prediction, while confidence levels obtained from frequentist methods provide more extensive coverage over data variations. Computationally, quantile regression type methods are cheaper for a single confidence interval but require re-training for different intervals. Sampling based methods generate samples that can form multiple confidence intervals, albeit at a higher computational cost.

cs.AI↗

arXiv2021

Deep Bayesian Active Learning for Accelerating Stochastic Simulation

Stochastic simulations such as large-scale, spatiotemporal, age-structured epidemic models are computationally expensive at fine-grained resolution. While deep surrogate models can speed up the simulations, doing so for stochastic simulations and with active learning approaches is an underexplored area. We propose Interactive Neural Process (INP), a deep Bayesian active learning framework for learning deep surrogate models to accelerate stochastic simulations. INP consists of two components, a spatiotemporal surrogate model built upon Neural Process (NP) family and an acquisition function for active learning. For surrogate modeling, we develop Spatiotemporal Neural Process (STNP) to mimic the simulator dynamics. For active learning, we propose a novel acquisition function, Latent Information Gain (LIG), calculated in the latent space of NP based models. We perform a theoretical analysis and demonstrate that LIG reduces sample complexity compared with random sampling in high dimensions. We also conduct empirical studies on three complex spatiotemporal simulators for reaction diffusion, heat flow, and infectious disease. The results demonstrate that STNP outperforms the baselines in the offline learning setting and LIG achieves the state-of-the-art for Bayesian active learning.

cs.LG↗

arXiv2021

DeepGLEAM: A hybrid mechanistic and deep learning model for COVID-19 forecasting

We introduce DeepGLEAM, a hybrid model for COVID-19 forecasting. DeepGLEAM combines a mechanistic stochastic simulation model GLEAM with deep learning. It uses deep learning to learn the correction terms from GLEAM, which leads to improved performance. We further integrate various uncertainty quantification methods to generate confidence intervals. We demonstrate DeepGLEAM on real-world COVID-19 mortality forecasting tasks.

cs.LG↗

arXiv2020

An Early Warning Approach to Monitor COVID-19 Activity with Multiple Digital Traces in Near Real-Time

Non-pharmaceutical interventions (NPIs) have been crucial in curbing COVID-19 in the United States (US). Consequently, relaxing NPIs through a phased re-opening of the US amid still-high levels of COVID-19 susceptibility could lead to new epidemic waves. This calls for a COVID-19 early warning system. Here we evaluate multiple digital data streams as early warning indicators of increasing or decreasing state-level US COVID-19 activity between January and June 2020. We estimate the timing of sharp changes in each data stream using a simple Bayesian model that calculates in near real-time the probability of exponential growth or decay. Analysis of COVID-19-related activity on social network microblogs, Internet searches, point-of-care medical software, and a metapopulation mechanistic model, as well as fever anomalies captured by smart thermometer networks, shows exponential growth roughly 2-3 weeks prior to comparable growth in confirmed COVID-19 cases and 3-4 weeks prior to comparable growth in COVID-19 deaths across the US over the last 6 months. We further observe exponential decay in confirmed cases and deaths 5-6 weeks after implementation of NPIs, as measured by anonymized and aggregated human mobility data from mobile phones. Finally, we propose a combined indicator for exponential growth in multiple data streams that may aid in developing an early warning system for future COVID-19 outbreaks. These efforts represent an initial exploratory framework, and both continued study of the predictive power of digital indicators as well as further development of the statistical approach are needed.

stat.AP↗

arXiv2020

Finding Patient Zero: Learning Contagion Source with Graph Neural Networks

Locating the source of an epidemic, or patient zero (P0), can provide critical insights into the infection's transmission course and allow efficient resource allocation. Existing methods use graph-theoretic centrality measures and expensive message-passing algorithms, requiring knowledge of the underlying dynamics and its parameters. In this paper, we revisit this problem using graph neural networks (GNNs) to learn P0. We establish a theoretical limit for the identification of P0 in a class of epidemic models. We evaluate our method against different epidemic models on both synthetic and a real-world contact network considering a disease with history and characteristics of COVID-19. % We observe that GNNs can identify P0 close to the theoretical bound on accuracy, without explicit input of dynamics or its parameters. In addition, GNN is over 100 times faster than classic methods for inference on arbitrary graph topologies. Our theoretical bound also shows that the epidemic is like a ticking clock, emphasizing the importance of early contact-tracing. We find a maximum time after which accurate recovery of the source becomes impossible, regardless of the algorithm used.

cs.SI↗

arXiv2020

A machine learning methodology for real-time forecasting of the 2019-2020 COVID-19 outbreak using Internet searches, news alerts, and estimates from mechanistic models

We present a timely and novel methodology that combines disease estimates from mechanistic models with digital traces, via interpretable machine-learning methodologies, to reliably forecast COVID-19 activity in Chinese provinces in real-time. Specifically, our method is able to produce stable and accurate forecasts 2 days ahead of current time, and uses as inputs (a) official health reports from Chinese Center Disease for Control and Prevention (China CDC), (b) COVID-19-related internet search activity from Baidu, (c) news media activity reported by Media Cloud, and (d) daily forecasts of COVID-19 activity from GLEAM, an agent-based mechanistic model. Our machine-learning methodology uses a clustering technique that enables the exploitation of geo-spatial synchronicities of COVID-19 activity across Chinese provinces, and a data augmentation technique to deal with the small number of historical disease activity observations, characteristic of emerging outbreaks. Our model's predictive power outperforms a collection of baseline models in 27 out of the 32 Chinese provinces, and could be easily extended to other geographies currently affected by the COVID-19 outbreak to help decision makers.

stat.OT↗

arXiv2020

Inferring high-resolution human mixing patterns for disease modeling

Mathematical and computational modeling approaches are increasingly used as quantitative tools in the analysis and forecasting of infectious disease epidemics. The growing need for realism in addressing complex public health questions is however calling for accurate models of the human contact patterns that govern the disease transmission processes. Here we present a data-driven approach to generate effective descriptions of population-level contact patterns by using highly detailed macro (census) and micro (survey) data on key socio-demographic features. We produce age-stratified contact matrices for 277 sub-national administrative regions of countries covering approximately 3.5 billion people and reflecting the high degree of cultural and societal diversity of the focus countries. We use the derived contact matrices to model the spread of airborne infectious diseases and show that sub-national heterogeneities in human mixing patterns have a marked impact on epidemic indicators such as the reproduction number and overall attack rate of epidemics of the same etiology. The contact patterns derived here are made publicly available as a modeling tool to study the impact of socio-economic differences and demographic heterogeneities across populations on the epidemiology of infectious diseases.

q-bio.PE↗

arXiv2019

The Emergence of Innovation Complexity at Different Geographical and Technological Scales

We define a novel quantitative strategy inspired by the ecological notion of nestedness to single out the scale at which innovation complexity emerges from the aggregation of specialized building blocks. Our analysis not only suggests that the innovation space can be interpreted as a natural system in which advantageous capabilities are selected by evolutionary pressure, but also that the emerging structure of capabilities is not independent of the scale of observation at which they are observed. Expanding on this insight allows us to understand whether the capabilities characterizing the innovation space at a given scale are compatible with a complex evolutionary dynamics or, rather, a set of essentially independent activities allowing to reduce the system at that scale to a set of disjoint non interacting sub-systems. This yields a measure of the innovation complexity of the system, i.e. of the degree of interdependence between the sets of capabilities underlying the system's building blocks.

econ.GN↗