A Cinema Challenge hackathon was held from 14 to 22 November of 2020. It was dedicated to creating solutions for online theatre sweet.tv. I managed to create a prize-winning solution for challenge 2 and decided to share my methodology. The task was to predict top-5 tv-programs for each user based on his view history. TV-program was considered watched if the user watched over 80% of it and did not change the channel.

Memes on the Internet are often harmless and sometimes amusing. However, by using certain types of images, text, or combinations of both, the seemingly harmless meme becomes a multimodal type of hate speech -- a hateful meme. The Hateful Memes Challenge is a first-of-its-kind competition which focuses on detecting hate speech in multimodal memes and it proposes a new data set containing 10,000+ new examples of multimodal content. We utilize VisualBERT -- which meant to be the BERT of vision and language -- that was trained multimodally on images and captions and apply Ensemble Learning. Our approach achieves 0.811 AUROC with an accuracy of 0.765 on the challenge test set and placed third out of 3,173 participants in the Hateful Memes Challenge.