Show Why the Answer is Correct! Towards Explainable AI using Compositional Temporal Attention

Abstract

Visual Question Answering (VQA) models have achieved significant success in recent times. Despite the success of VQA models, they are mostly black-box models providing no reasoning about the predicted answer, thus raising questions for their applicability in safety-critical such as autonomous systems and cyber-security. Current state of the art fail to better complex questions and thus are unable to exploit compositionality. To minimize the black-box effect of these models and also to make them better exploit compositionality, we propose a Dynamic Neural Network (DMN), which can understand a particular question and then dynamically assemble various relatively shallow deep learning modules from a pool of modules to form a network. We incorporate compositional temporal attention to these deep learning based modules to increase compositionality exploitation. This results in achieving better understanding of complex questions and also provides reasoning as to why the module predicts a particular answer. Experimental analysis on the two benchmark datasets, VQA2.0 and CLEVR, depicts that our model outperforms the previous approaches for Visual Question Answering task as well as provides better reasoning, thus making it reliable for mission critical applications like safety and security.

Deep Learning: Historical Overview from Inception to Actualization, Models, Applications and Future Trends

“…Layer-wise relevance propagation involves the decomposition of the model's prediction by assigning importance score to features in each layer of a neural network and also providing information about what each component has contributed to the model's final predictions (Wenli et al, 2023). Attention mechanisms worked with the model that implements attention mechanism, and it visualises the learning weight of the model to figure out the aspect of the input sequence that the model considered for its predictions (Bendre et al, 2021;Ntrougkas et al, 2022;Gkartzonika et al, 2023).…”

Using biologically hierarchical modular architecture for explainable, tunable, generalizable, spatial AI

“…Modular hierarchical or collaborative ANN models can be more explainable due to their organization. 22,23 While not completely transparent, hierarchical ensemble ANN models might advance the fields of computational neuroscience and applied AI by enabling the design of AI platforms that are transparent enough for ethical examination and functional tuning without losing the strengths of data-based code generation that characterizes ML. BHMA models (figure 2) use context-dependent integration ANNs for executive control of encapsulated purpose-tailored ANN submodules.…”