Next-generation deep learning based on simulators and synthetic data

CM de Melo, A Torralba, L Guibas, J DiCarlo… - Trends in cognitive …, 2022 - cell.com
Deep learning (DL) is being successfully applied across multiple domains, yet these models
learn in a most artificial way: they require large quantities of labeled data to grasp even …

Incorporating physics into data-driven computer vision

A Kadambi, C de Melo, CJ Hsieh… - Nature Machine …, 2023 - nature.com
Many computer vision techniques infer properties of our physical world from images.
Although images are formed through the physics of light and mechanics, computer vision …

Integration of neural network-based symbolic regression in deep learning for scientific discovery

S Kim, PY Lu, S Mukherjee, M Gilbert… - IEEE transactions on …, 2020 - ieeexplore.ieee.org
Symbolic regression is a powerful technique to discover analytic equations that describe
data, which can lead to explainable models and the ability to predict unseen data. In …

Synthetic data in healthcare

D McDuff, T Curran, A Kadambi - arXiv preprint arXiv:2304.03243, 2023 - arxiv.org
Synthetic data are becoming a critical tool for building artificially intelligent systems.
Simulators provide a way of generating data systematically and at scale. These data can …

Intelligent computational techniques for physical object properties discovery, detection, and prediction: A comprehensive survey

S Mishra, A Arora - Computer Science Review, 2024 - Elsevier
The exploding usage of physical object properties has greatly facilitated real-time
applications such as robotics to perceive exactly as it appears in existence. Changes in the …

Neural implicit representations for physical parameter inference from a single video

F Hofherr, L Koestler, F Bernard… - Proceedings of the …, 2023 - openaccess.thecvf.com
Neural networks have recently been used to analyze diverse physical systems and to
identify the underlying dynamics. While existing methods achieve impressive results, they …

Physics-AI symbiosis

B Jalali, Y Zhou, A Kadambi… - … Learning: Science and …, 2022 - iopscience.iop.org
The phenomenal success of physics in explaining nature and engineering machines is
predicated on low dimensional deterministic models that accurately describe a wide range …

Neural Material Adaptor for Visual Grounding of Intrinsic Dynamics

J Cao, S Guan, Y Ge, W Li, X Yang, C Ma - arXiv preprint arXiv …, 2024 - arxiv.org
While humans effortlessly discern intrinsic dynamics and adapt to new scenarios, modern AI
systems often struggle. Current methods for visual grounding of dynamics either use pure …

Methodology Development of a Free-Flight Parameter Estimation Technique Using Physics-Informed Neural Networks

N Michek, P Mehta, W Huebsch - 2023 IEEE Aerospace …, 2023 - ieeexplore.ieee.org
Unstable free-flight rigid body motion, consisting of 3D translational motion and large
angular rates about all axes and orientations outside of typical flight envelopes, is a complex …

Blending physics with artificial intelligence

A Kadambi - Computational Imaging V, 2020 - spiedigitallibrary.org
For centuries, humans have discovered the physical laws that underpin our world. What if
the next Einstein or Newton is not a human, but a machine? Machines that are physics …