More Like this

1. Slowing the body slows down time perception

   https://doi.org/10.7554/eLife.63607  

   De Kock, Rose; Zhou, Weiwei; Joiner, Wilsaan M; Wiener, Martin (April 2021, eLife)

   null (Ed.)

   Interval timing is a fundamental component of action and is susceptible to motor-related temporal distortions. Previous studies have shown that concurrent movement biases temporal estimates, but have primarily considered self-modulated movement only. However, real-world encounters often include situations in which movement is restricted or perturbed by environmental factors. In the following experiments, we introduced viscous movement environments to externally modulate movement and investigated the resulting effects on temporal perception. In two separate tasks, participants timed auditory intervals while moving a robotic arm that randomly applied four levels of viscosity. Results demonstrated that higher viscosity led to shorter perceived durations. Using a drift-diffusion model and a Bayesian observer model, we confirmed these biasing effects arose from perceptual mechanisms, instead of biases in decision making. These findings suggest that environmental perturbations are an important factor in movement-related temporal distortions, and enhance the current understanding of the interactions of motor activity and cognitive processes. 

   more » « less
2. Improving Robustness of Convolutional Networks Through Sleep-Like Replay

   https://doi.org/10.1109/ICMLA58977.2023.00043  

   Delanois, Jean Erik; Ahuja, Aditya; Krishnan, Giri P; Tadros, Timothy; McAuley, Julian; Bazhenov, Maxim (December 2023, IEEE)

   Convolutional neural networks (CNNs) are a foundational model architecture utilized to perform a wide variety of visual tasks. On image classification tasks CNNs achieve high performance, however model accuracy degrades quickly when inputs are perturbed by distortions such as additive noise or blurring. This drop in performance partly arises from incorrect detection of local features by convolutional layers. In this work, we develop a neuroscience-inspired unsupervised Sleep Replay Consolidation (SRC) algorithm for improving convolutional filter’s robustness to perturbations. We demonstrate that sleep- based optimization improves the quality of convolutional layers by the selective modification of spatial gradients across filters. We further show that, compared to other approaches such as fine- tuning, a single sleep phase improves robustness across different types of distortions in a data efficient manner. 

   more » « less
3. ShapeShifter: Robust Physical Adversarial Attack on Faster R-CNN Object Detector

   https://doi.org/10.1007/978-3-030-10925-7_4  

   Chen, Shang-Tse; Cornelius, Cory; Martin, Jason; Chau, Duen Horng (September 2018, Joint European Conference on Machine Learning and Knowledge Discovery in Databases)

   Given the ability to directly manipulate image pixels in the digital input space, an adversary can easily generate imperceptible perturbations to fool a Deep Neural Network (DNN) image classifier, as demonstrated in prior work. In this work, we propose ShapeShifter, an attack that tackles the more challenging problem of crafting physical adversarial perturbations to fool image-based object detectors like Faster R-CNN. Attacking an object detector is more difficult than attacking an image classifier, as it needs to mislead the classification results in multiple bounding boxes with different scales. Extending the digital attack to the physical world adds another layer of difficulty, because it requires the perturbation to be robust enough to survive real-world distortions due to different viewing distances and angles, lighting conditions, and camera limitations. We show that the Expectation over Transformation technique, which was originally proposed to enhance the robustness of adversarial perturbations in image classification, can be successfully adapted to the object detection setting. ShapeShifter can generate adversarially perturbed stop signs that are consistently mis-detected by Faster R-CNN as other objects, posing a potential threat to autonomous vehicles and other safety-critical computer vision systems. 

   more » « less
4. Fidelity of low-frequency underwater acoustic measurements by sensors mounted on compact platforms

   https://doi.org/10.1121/1.5130755  

   Godin, Oleg_A (November 2019, The Journal of the Acoustical Society of America)

   Measurements by sensors mounted on compact platforms are affected by sound scattering from the platform. Assuming a spherical shape of the platform, this paper investigates the differences between the ambient and measured characteristics of low-frequency signals and noise for scalar and vector sensors. In the near field of the platform, low-frequency perturbations in oscillatory velocity are generally much larger than pressure perturbations. These perturbations prevent mounted vector sensors from correctly measuring the direction of the free-field oscillatory velocity. The feasibility of a compensation of the distortions in scalar and vector sensor measurements is discussed. 

   more » « less
5. Scaffolding Simulations with Deep Learning for High-Dimensional Deconvolution

   Andreassen, Anders; Komiske, Patrick T.; Metodiev, Eric. M.; Nachman, Benjamin; Suresh, Adi; Thaler, Jesse (April 2021, ICLR 2021 SimDL Workshop)

   null (Ed.)

   A common setting for scientific inference is the ability to sample from a high-fidelity forward model (simulation) without having an explicit probability density of the data. We propose a simulation-based maximum likelihood deconvolution approach in this setting called OMNIFOLD. Deep learning enables this approach to be naturally unbinned and (variable-, and) high-dimensional. In contrast to model parameter estimation, the goal of deconvolution is to remove detector distortions in order to enable a variety of down-stream inference tasks. Our approach is the deep learning generalization of the common Richardson-Lucy approach that is also called Iterative Bayesian Unfolding in particle physics. We show how OMNIFOLD can not only remove detector distortions, but it can also account for noise processes and acceptance effects. 

   more » « less