How do you measure trust in deep learning?

23 January 2021 05:00 PM

Whether it’s diagnosing patients or driving cars, we want to know whether we can trust a person before assigning them a sensitive task. In the human world, we have different ways to establish and measure trustworthiness. In artificial intelligence, the establishment of trust is still developing. (via @reckoningdev)

Trust in Deep Learning

End-to-end object detection with Transformers

4 January 2021 12:00 AM

Traditional computer vision models typically use a complex, partly handcrafted pipeline that relies on custom layers in order to localize objects in an image and then extract features. DETR replaces this with a simpler neural network that offers a true end-to-end deep learning solution to the problem. (via @reckoningdev)

Facebook AI Blog

Data-efficient image Transformers

4 January 2021 12:00 AM

Image classification — the task of understanding the main content of an image — is easy for humans but hard for machines. In particular, it is challenging for convolution-free Transformers like DeiT because these systems don’t have many statistical priors about images: They typically have to “see” a lot of example images in order to learn to classify different objects. DeiT, however, can be trained effectively with 1.2 million images, rather than requiring hundreds of millions of images. (via @reckoningdev)

Facebook AI Blog

The Surprising Effectiveness of Linear Unsupervised Image-to-Image Translation

7 August 2020 10:44 PM

Unsupervised image-to-image translation is an inherently ill-posed problem. This paper uses linear encoder-decoder architectures for unsupervised image to image translation. (via @reckoningdev)

Robust Training with Ensemble Consensus

2 August 2020 08:53 PM

Since deep neural networks are over-parameterized, they can memorize noisy examples. We address such memorizing issue in the presence of annotation noise. (via @reckoningdev)

BorderDet: Border Feature for Dense Object Detection

1 August 2020 10:37 PM

Dense object detectors rely on the sliding-window paradigm that predicts the object over a regular grid of image. Meanwhile, the feature maps on the point of the grid are adopted to generate… (via @reckoningdev)