Faster-LTN: a neuro-symbolic, end-to-end object detection architecture
This addresses the challenge of integrating semantic reasoning with neural learning for object detection, but it is incremental as it builds on existing neuro-symbolic and detection frameworks.
The paper tackles the problem of detecting semantic relationships between objects in images by proposing Faster-LTN, a neuro-symbolic architecture combining a convolutional backbone with Logic Tensor Networks for end-to-end training. It achieves competitive performance compared to Faster R-CNN.
The detection of semantic relationships between objects represented in an image is one of the fundamental challenges in image interpretation. Neural-Symbolic techniques, such as Logic Tensor Networks (LTNs), allow the combination of semantic knowledge representation and reasoning with the ability to efficiently learn from examples typical of neural networks. We here propose Faster-LTN, an object detector composed of a convolutional backbone and an LTN. To the best of our knowledge, this is the first attempt to combine both frameworks in an end-to-end training setting. This architecture is trained by optimizing a grounded theory which combines labelled examples with prior knowledge, in the form of logical axioms. Experimental comparisons show competitive performance with respect to the traditional Faster R-CNN architecture.