My personal blog

Machine learning, computer vision, languages

Sigmoid activation is not optimal with binary segmentation
05 September 2021
The standard activation function for binary outputs is the sigmoid function. However, in a recent paper, I show empirically on several medical segmentation datasets that other functions can be better.

Operations on contextual word embeddings
16 June 2021
A variety of operations can be performed on (contextual) word vectors. In this blog post, I will implement some common operations using PyTorch and Python.

Uncertainty estimation in neural networks
14 August 2020
In this blog post, I will implement some common methods for uncertainty estimation. My main focus lies on classification and segmentation. Therefore, regression-specific methods such as Pinball loss are not covered here.

Metrics for uncertainty estimation
07 August 2020
Predictions are not just about accuracy, but also about probability. In lots of applications it is important to know how sure a neural network is of a prediction. However, the softmax probabilities in neural networks are not always calibrated and don’t necessarily measure uncertainty. In this blog post, I will implement the most common metrics to evaluate the output probabilities of neural networks.

Implementing Poincaré Embeddings in PyTorch
24 July 2020
After having introduced Riemannian SGD in the last blog post, here I will give a concrete application for this optimization method. Poincaré embeddings [1][2] are hierarchical word embeddings which map integer-encoded words to the hyperbolic space.

Older Posts →