Collaborating Authors


DeviantArt Is Using AI To Alert Artists When Their Work Is Stolen For NFTs


Art theft has become a major problem in the world of Non-Fungible Tokens (NFTs) as grifters look to make a quick buck from the works of others. The nature of the online goods means it's very difficult to confirm who owns the NFTs being sold and if the sellers have the legal right to sell that work on any platform. Progress on a solution has been slow, but it does appear new tactics from hosting companies like DeviantArt are working. DeviantArt recently implemented a new system designed to help identify stolen artwork in the wild by using machine learning to locate works that may have been stolen. It's even able to detect subtle variations in stolen artwork, including if an image is cropped, flipped or slightly altered to avoid traditional image detection systems.

Artists Are Now Using AI to Prevent Their Work From Being Stolen, Sold as NFTs


From attaching value to paper to make currency notes to attaching value to digital art works that sometimes come in the form of jpegs, the world economy has come a long way. Non-fungible tokens (NFTs) are the new currency for the digital users who invest in cryptocurrency. However, along with the NFT boom, it has also been reported that several artworks are being stolen and sold on NFT marketplaces. This has left digital creators in a fix as their one opportunity to earn money from their work has come under threat. Platforms like DeviantArt, an American online art community featuring artwork, videography and photography have now become a place for online burglars to steal the pieces created by artists which they then sell on NFT platforms.

Recommending Groups to Users Using User-Group Engagement and Time-Dependent Matrix Factorization

AAAI Conferences

Social networks often provide group features to help users with similar interests associate and consume content together. Recommending groups to users poses challenges due to their complex relationship: user-group affinity is typically measured implicitly and varies with time; similarly, group characteristics change as users join and leave. To tackle these challenges, we adapt existing matrix factorization techniques to learn user-group affinity based on two different implicit engagement metrics: (i) which group-provided content users consume; and (ii) which content users provide to groups. To capture the temporally extended nature of group engagement we implement a time-varying factorization. We test the assertion that latent preferences for groups and users are sparse in investigating elastic-net regularization. Experiments using data from DeviantArt indicate that the time-varying implicit engagement-based model provides the best top-K group recommendations, illustrating the benefit of the added model complexity.