Attending to Mathematical Language with Transformers

5 Dec 20181 code implementation

Mathematical expressions were generated, evaluated and used to train neural network models based on the transformer architecture.

Dynamic Routing Between Capsules

NeurIPS 2017 39 code implementations

We use the length of the activity vector to represent the probability that the entity exists and its orientation to represent the instantiation parameters.


Digital phase-only holography using deep conditional generative models

3 Nov 20191 code implementation

Holographic wave-shaping has found numerous applications across the physical sciences, especially since the development of digital spatial-light modulators (SLMs).

Recurrent Spatial Transformer Networks

17 Sep 20151 code implementation

We investigate different down-sampling factors (ratio of pixel in input and output) for the SPN and show that the RNN-SPN model is able to down-sample the input images without deteriorating performance.

DDSP: Differentiable Digital Signal Processing

14 Jan 20201 code implementation

In this paper, we introduce the Differentiable Digital Signal Processing (DDSP) library, which enables direct integration of classic signal processing elements with deep learning methods.


ShapeShifter: Robust Physical Adversarial Attack on Faster R-CNN Object Detector

16 Apr 20181 code implementation

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.


Learning sparse transformations through backpropagation

22 Oct 20182 code implementations

Many transformations in deep learning architectures are sparsely connected.

Supervised Multimodal Bitransformers for Classifying Images and Text

6 Sep 20191 code implementation

Self-supervised bidirectional transformer models such as BERT have led to dramatic improvements in a wide variety of textual classification tasks.

Parallel fault-tolerant programming of an arbitrary feedforward photonic network

11 Sep 20191 code implementation

Reconfigurable photonic mesh networks of tunable beamsplitter nodes can linearly transform $N$-dimensional vectors representing input modal amplitudes of light for applications such as energy-efficient machine learning hardware, quantum information processing, and mode demultiplexing.


GAN-based Virtual Re-Staining: A Promising Solution for Whole Slide Image Analysis

13 Jan 20191 code implementation

We proposed a conditional CycleGAN (cCGAN) network to transform the H\&E stained images into IHC stained images, facilitating virtual IHC staining on the same slide.