Search Results for author: Josue Ortega Caro

Found 9 papers, 4 papers with code

Linking convolutional kernel size to generalization bias in face analysis CNNs

no code implementations7 Feb 2023 Hao Liang, Josue Ortega Caro, Vikram Maheshri, Ankit B. Patel, Guha Balakrishnan

Our framework is experimental, in that we train several versions of a network with an intervention to a specific hyperparameter, and measure the resulting causal effect of this choice on performance bias when a particular out-of-distribution image perturbation is applied.

Continuous Spatiotemporal Transformers

1 code implementation31 Jan 2023 Antonio H. de O. Fonseca, Emanuele Zappala, Josue Ortega Caro, David van Dijk

Modeling spatiotemporal dynamical systems is a fundamental challenge in machine learning.

Neural Integral Equations

1 code implementation30 Sep 2022 Emanuele Zappala, Antonio Henrique de Oliveira Fonseca, Josue Ortega Caro, David van Dijk

In this paper, we introduce Neural Integral Equations (NIE), a method that learns an unknown integral operator from data through an IE solver.

Understanding robustness and generalization of artificial neural networks through Fourier masks

1 code implementation16 Mar 2022 Nikos Karantzas, Emma Besier, Josue Ortega Caro, Xaq Pitkow, Andreas S. Tolias, Ankit B. Patel, Fabio Anselmi

Our results also indicate that the essential frequencies in question are effectively the ones used to achieve generalization in the first place.

Data Augmentation

Shallow Univariate ReLu Networks as Splines: Initialization, Loss Surface, Hessian, & Gradient Flow Dynamics

no code implementations4 Aug 2020 Justin Sahs, Ryan Pyle, Aneel Damaraju, Josue Ortega Caro, Onur Tavaslioglu, Andy Lu, Ankit Patel

Our implicit regularization results are complementary to recent work arXiv:1906. 07842, done independently, which showed that initialization scale critically controls implicit regularization via a kernel-based argument.

Inductive Bias

Local Convolutions Cause an Implicit Bias towards High Frequency Adversarial Examples

no code implementations19 Jun 2020 Josue Ortega Caro, Yilong Ju, Ryan Pyle, Sourav Dey, Wieland Brendel, Fabio Anselmi, Ankit Patel

Inspired by theoretical work on linear full-width convolutional models, we hypothesize that the local (i. e. bounded-width) convolutional operations commonly used in current neural networks are implicitly biased to learn high frequency features, and that this is one of the root causes of high frequency adversarial examples.

Adversarial Robustness Vocal Bursts Intensity Prediction

A Functional Characterization of Randomly Initialized Gradient Descent in Deep ReLU Networks

no code implementations25 Sep 2019 Justin Sahs, Aneel Damaraju, Ryan Pyle, Onur Tavaslioglu, Josue Ortega Caro, Hao Yang Lu, Ankit Patel

Despite their popularity and successes, deep neural networks are poorly understood theoretically and treated as 'black box' systems.

Recurrent computations for visual pattern completion

1 code implementation7 Jun 2017 Hanlin Tang, Martin Schrimpf, Bill Lotter, Charlotte Moerman, Ana Paredes, Josue Ortega Caro, Walter Hardesty, David Cox, Gabriel Kreiman

First, subjects robustly recognized objects even when rendered <15% visible, but recognition was largely impaired when processing was interrupted by backward masking.

Image Classification

Cannot find the paper you are looking for? You can Submit a new open access paper.