We explore the hypothesis that poor compositional generalization in neural networks is caused by difficulties with learning effective routing. To solve this problem, we propose the concept of block-operations, which is based on splitting all activation tensors in the network into uniformly sized blocks and using an inductive bias to encourage modular routing and modification of these blocks. Based on this concept we introduce the Multiplexer, a new architectural component that enhances the Feed Forward Neural Network (FNN). We experimentally confirm that Multiplexers exhibit strong compositional generalization. On both a synthetic and a realistic task our model was able to learn the underlying process behind the task, whereas both FNNs and Transformers were only able to learn heuristic approximations. We propose as future work to use the principles of block-operations to improve other existing architectures.

The European Commission's Joint Research Center is working on a tool that could use tweets and artificial intelligence to collect real-time data on floods. In a paper released on Arvix.org, EU scientists explain how their Social Media for Flood Risk (SMFR) prototype could help emergency responders better understand what's happening on the ground in flooded areas and determine what trouble spots might need immediate attention. The tool works in collaboration with Europe's Flood Awareness System (EFAS). When EFAS identifies areas with heightened flood risks, it triggers SMFR to begin collecting flood-related tweets from users in those areas.