Claudio Silva
New York University · Computer Science
Active 1994–2024
About
My research interests are broad and inherently interdisciplinary. Within computer science, I focus on visualization, visual analytics, machine learning, reproducibility and provenance, geometric computing, urban computing, computer graphics, and computer vision. I particularly enjoy working at the intersection of different fields, recently collaborating with paleontologists, urban scientists, and sports scientists. I enjoy building tools that others can use to advance their own research and solve real-world problems.
Research topics
- Computer Science
- Artificial Intelligence
- Engineering
- Data science
- Machine Learning
- Cartography
- Geography
- Civil engineering
- Astrobiology
- Meteorology
- Speech recognition
- Physics
- Acoustics
- Software engineering
- Transport engineering
- Remote sensing
- Programming language
- Human–computer interaction
- Multimedia
- Geology
- Architectural engineering
- Telecommunications
Selected publications
CitySurfaces: City-scale semantic segmentation of sidewalk materials
Sustainable Cities and Society · 2022 · 46 citations
Senior authorCorresponding- Computer Science
- Artificial Intelligence
- Computer Science
A Study on Robustness to Perturbations for Representations of Environmental Sound
2021 29th European Signal Processing Conference (EUSIPCO) · 2022 · 7 citations
- Computer Science
- Computer Science
- Artificial Intelligence
Audio applications involving environmental sound analysis increasingly use general-purpose audio representations, also known as embeddings, for transfer learning. Recently, Holistic Evaluation of Audio Representations (HEAR) evaluated twenty-nine embedding models on nineteen diverse tasks. However, the evaluation's effectiveness depends on the variation already captured within a given dataset. Therefore, for a given data domain, it is unclear how the representations would be affected by the variations caused by myriad microphones' range and acoustic conditions – commonly known as channel effects. We aim to extend HEAR to evaluate invariance to channel effects in this work. To accomplish this, we imitate channel effects by injecting perturbations to the audio signal and measure the shift in the new (perturbed) embeddings with three distance measures, making the evaluation domain-dependent but not task-dependent. Combined with the downstream performance, it helps us make a more informed prediction of how robust the embeddings are to the channel effects. We evaluate two embeddings – YAMNet, and OpenL <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">3</sup> on monophonic (UrbanSound8K) and polyphonic (SONYC-UST) urban datasets. We show that one distance measure does not suffice in such task-independent evaluation. Although Fréchet Audio Distance (FAD) correlates with the trend of the performance drop in the downstream task most accurately, we show that we need to study FAD in conjunction with the other distances to get a clear understanding of the overall effect of the perturbation. In terms of the embedding performance, we find OpenL <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">3</sup> to be more robust than YAMNet, which aligns with the HEAR evaluation.
2020 · 27 citations
Senior authorCorresponding- Computer Science
- Computer Science
- Geography
Urban planning is increasingly data driven, yet the challenge of designing with data at a city scale and remaining sensitive to the impact at a human scale is as important today as it was for Jane Jacobs. We address this challenge with Urban Mosaic,a tool for exploring the urban fabric through a spatially and temporally dense data set of 7.7 million street-level images from New York City, captured over the period of a year. Working in collaboration with professional practitioners, we use Urban Mosaic to investigate questions of accessibility and mobility, and preservation and retrofitting. In doing so, we demonstrate how tools such as this might provide a bridge between the city and the street, by supporting activities such as visual comparison of geographically distant neighborhoods,and temporal analysis of unfolding urban development.
Interactive Visualization of Atmospheric Effects for Celestial Bodies
IEEE Transactions on Visualization and Computer Graphics · 2020 · 16 citations
Senior authorCorresponding- Computer Science
- Computer Science
- Artificial Intelligence
We present an atmospheric model tailored for the interactive visualization of planetary surfaces. As the exploration of the solar system is progressing with increasingly accurate missions and instruments, the faithful visualization of planetary environments is gaining increasing interest in space research, mission planning, and science communication and education. Atmospheric effects are crucial in data analysis and to provide contextual information for planetary data. Our model correctly accounts for the non-linear path of the light inside the atmosphere (in Earth's case), the light absorption effects by molecules and dust particles, such as the ozone layer and the Martian dust, and a wavelength-dependent phase function for Mie scattering. The mode focuses on interactivity, versatility, and customization, and a comprehensive set of interactive controls make it possible to adapt its appearance dynamically. We demonstrate our results using Earth and Mars as examples. However, it can be readily adapted for the exploration of other atmospheres found on, for example, of exoplanets. For Earth's atmosphere, we visually compare our results with pictures taken from the International Space Station and against the CIE clear sky model. The Martian atmosphere is reproduced based on available scientific data, feedback from domain experts, and is compared to images taken by the Curiosity rover. The work presented here has been implemented in the OpenSpace system, which enables interactive parameter setting and real-time feedback visualization targeting presentations in a wide range of environments, from immersive dome theaters to virtual reality headsets.
<i>PipelineProfiler:</i> A Visual Analytics Tool for the Exploration of AutoML Pipelines
IEEE Transactions on Visualization and Computer Graphics · 2020 · 48 citations
Senior authorCorresponding- Computer Science
- Machine Learning
- Computer Science
In recent years, a wide variety of automated machine learning (AutoML) methods have been proposed to generate end-to-end ML pipelines. While these techniques facilitate the creation of models, given their black-box nature, the complexity of the underlying algorithms, and the large number of pipelines they derive, they are difficult for developers to debug. It is also challenging for machine learning experts to select an AutoML system that is well suited for a given problem. In this paper, we present the Pipeline Profiler, an interactive visualization tool that allows the exploration and comparison of the solution space of machine learning (ML) pipelines produced by AutoML systems. PipelineProfiler is integrated with Jupyter Notebook and can be combined with common data science tools to enable a rich set of analyses of the ML pipelines, providing users a better understanding of the algorithms that generated them as well as insights into how they can be improved. We demonstrate the utility of our tool through use cases where PipelineProfiler is used to better understand and improve a real-world AutoML system. Furthermore, we validate our approach by presenting a detailed analysis of a think-aloud experiment with six data scientists who develop and evaluate AutoML tools.
Recent grants
U.S. Brazil Collaborative Research: 3D Modeling and Visualization
NSF · $85k · 2004–2009
NSF · $516k · 2008–2011
NSF · $800k · 2012–2017
MRI: Development of Reconfigurable Environmental Intelligence Platform
NSF · $600k · 2018–2023
Where the Ocean Meets the Cloud: Ad Hoc Longitudinal Analysis of Massive Mesh Data
NSF · $190k · 2009–2011
Frequent coauthors
- 135 shared
Hanspeter Pfister
Harvard University
- 134 shared
James Ahrens
Los Alamos National Laboratory
- 132 shared
Klaus Mueller
- 131 shared
Robert Moorhead
- 130 shared
Vice Chair
University of Utah
- 130 shared
Amitabh Varshney
University of Maryland, College Park
- 130 shared
Miriah Meyer
- 129 shared
Aditi Majumder
University of California, Irvine
Labs
Education
- 1990
B.S., Mathematics
Universidade Federal do Ceará (Brazil)
- 1996
M.S., Computer Science
State University of New York at Stony Brook
- 1996
Ph.D., Computer Science
State University of New York at Stony Brook
Awards & honors
- Fellow of the ACM (2024)
- Fellow of IEEE (2013)
- IEEE Visualization Technical Achievement Award (2014)
Similar researchers at New York University
- Resume-aware match score
- Save to shortlist
- AI-drafted outreach
See your match with Claudio Silva
PhdFit ranks faculty by your research interests, methods, and publications — grounded in their actual work, not templates.
- Free to start
- No credit card
- 30-second signup