Newsroom
Latest news from the team at Raffle
Autoencoding beyond pixels using a learned similarity metric
We present an autoencoder that leverages learned representations to better measure similarities in data space.
Explore our Research By combining a variational autoencoder with a generative adversarial network, we can use learned feature representations in the GAN discriminator as a basis for the VAE reconstruction objective. We replace element-wise errors with feature-wise errors to better capture the data distribution while offering invariance towards, e.g., translation. We apply our method to images of faces and show that it outperfo…
Jul 26, 2022
English
Webinar: AI and its role in business
Watch a discussion between Raffles former CEO and CRO about recent megatrends in AI and opportunities for companies to expand.
Watch our Webinar Even though over 60% of leading global companies have AI on the agenda, only 4% are actively using it. The thing is, getting AI right is often harder than one initially thinks. Suzanne Lauritzen and Ole Winther discuss the recent megatrends in AI and how companies can get on board with them. Watch the AI Talks webinar recording to find out how one of the leading global experts in AI sees the future.
Jul 26, 2022
English
The science behind the Raffle-lution: Natural language search
What is NLP? Read more about natural language processing within Raffle and how the software understands contextual meaning from text.
This is the second in our “The science behind the Raffle-lution” series. Read the first here . When we think about lightning-fast search capabilities today, we think about the masters: Google. Using their search engine, you can search the entire internet at the click of a button. How did they achieve this? Search engines represent text in any document (or web page) with an index which in its simplest form is a list of all the…
Jul 26, 2022
Product
English
Semi-supervised generation with cluster-aware generative models
Deep generative models trained with large amounts of unlabelled data have proven to be powerful within the domain of unsupervised learning.
Many real-life data sets contain a small number of labeled data points that are typically disregarded when training generative models. We propose the Cluster-aware Generative Model that uses unlabelled information to infer a latent representation that models the natural clustering of the data and additional labeled data points to refine this clustering. The generative performances of the model significantly improve when labele…
Jul 26, 2022
English
