This hybrid AI system can understand causality in controlled environments

In their paper, the researchers introduce CLEVRER, a new dataset and benchmark to evaluate the capabilities of AI algorithms in reasoning about video sequences, and Neuro-Symbolic Dynamic Reasoning (NS-DR), a hybrid AI system that marks a substantial improvement on causal reasoning in controlled environments.Read: [Microsoft’s new AI can generate smart to-do lists from your emails]For us humans, detecting and reasoning about objects in a scene almost go hand in hand. The AI agent must be able to parse the scene and answer multichoice questions about the number of objects, their attributes, and their spatial relationships.CLEVR is a visual question-answering dataset that tests the capabilities of AI systems in reasoning about the content of images. But they are complicated tasks to accomplish with current blends of AI because they require a causal understanding of the scene.As the authors of the paper summarize, solving CLEVRER problems requires three key elements: “recognition of the objects and events in the videos; modeling the dynamics and causal relations between the objects and events; and understanding of the symbolic logic behind the questions.”“CLEVRER is a first visual reasoning dataset that is designed for casual reasoning in videos. They also lack a model of the world that allows them to foresee what happens next and figure out how alternative counterfactual scenarios work.As a solution, the researchers introduced the Neuro-Symbolic Dynamic Reasoning model, a combination of neural networks and symbolic artificial intelligence. But on the other hand, rule-based systems are very good at symbolic reasoning and knowledge representation, an area that has been a historical pain point for machine learning algorithms.NS-DR puts both neural networks and symbolic reasoning systems to good use:The Neuro-Symbolic Dynamic Reasoning model puts together neural networks and symbolic artificial intelligenceThe performance of NS-DR is considerably higher than pure deep learning models on explanatory, predictive, and counterfactual challenges. But it is still a significant gain in comparison to the 25-percent accuracy of the best-performing baseline deep learning model.Another significant benefit of NS-DR is that it requires much less data in the training phase.The results show that incorporating neural networks and symbolic programs in the same AI model can combine their strengths and overcome their weaknesses.