Data, code & materials from the paper "Generalisation in humans and deep neural networks" (NeurIPS 2018)

Figures & code from the paper "Shortcut Learning in Deep Neural Networks" (Nature Machine Intelligence 2020)

A spreadsheet-like interactive evaluator for F#

The official repository for our paper "The Neural Data Router: Adaptive Control Flow in Transformers Improves Systematic Generalization".

Code from the article: "The Role of Disentanglement in Generalisation" (ICLR, 2021).

Manipulation of geographical and statistical data in Java, with a focus on Eurostat data

Label smoothed Aggregation cross entropy loss for generalisation in sequence to sequence tasks.

In the context of Deep Learning: What is the right way to conduct example weighting? How do you understand loss functions and so-called theorems on them?

Regularize polygons to rectilinear shape using constrained least squares in Java

Code from the article: "Lost in Latent Space: Examining Failures of Disentangled Models at Combinatorial Generalisaton" (NeurIPS, 2022)

Investigation of how noise perturbations impact neural network calibration and generalisation

This is a simple DAO for Java with some more features. I used it for my school project.

Metrics to assess the generalisation ability of NILM algorithms

LatentAugment, is a new data augmentation method for training any deep models. It navigates GAN's latent space to increase the diversity and quality of generated samples and enhance their effectiveness for DA purposes.

Generalisation of electrical models of neurons with MCMC

Web-based dot map generalisation

Supervised Learning - Regression Algorithm

Framework for benchmarking black-box adversarial attacks, modeled with ES.

Polynomials are mathematical objects which are expressions of variables and coefficients added to-gether. These polynomials can use operator overloading in a very intuitive manner since most of the operators that we can overload are mathematical operators, which can also be applied to polynomials.