Deep neural networks have revolutionized numerous research fields and applications. Despite their widespread success, a fundamental limitation known as catastrophic forgetting remains, where models fail to retain their ability to perform previously learned tasks after being trained on new ones. This limitation is particularly acute in certain continual learning scenarios, where models must integrate the knowledge from new domains with their existing capabilities. Traditional approaches to mitigate this problem typically rely on memory replay mechanisms, storing either original data samples, prototypes, or activation patterns. Although effective, these methods often introduce significant computational overhead, raise privacy concerns, and require the use of dedicated architectures. In this work we present ReReLRP (Remembering and Recognizing with LRP), a novel solution that leverages Layerwise Relevance Propagation (LRP) to preserve information across tasks. Our contribution provides increased privacy of existing replay-free methods while additionally offering built-in explainability, flexibility of model architecture and deployment, and a new mechanism to increase memory storage efficiency. We validate our approach on a wide variety of datasets, demonstrating results comparable with a well-known replay-based method in selected scenarios.

Recent research has revealed that deep neural networks often take dataset biases as a shortcut to make decisions rather than understand tasks, leading to failures in real-world applications. In this study, we focus on the spurious correlation between word features and labels that models learn from the biased data distribution of training data. In particular, we define the word highly co-occurring with a specific label as biased word, and the example containing biased word as biased example. Our analysis shows that biased examples are easier for models to learn, while at the time of prediction, biased words make a significantly higher contribution to the models' predictions, and models tend to assign predicted labels over-relying on the spurious correlation between words and labels. To mitigate models' over-reliance on the shortcut (i.e. spurious correlation), we propose a training strategy Less-Learn-Shortcut (LLS): our strategy quantifies the biased degree of the biased examples and down-weights them accordingly. Experimental results on Question Matching, Natural Language Inference and Sentiment Analysis tasks show that LLS is a task-agnostic strategy and can improve the model performance on adversarial data while maintaining good performance on in-domain data.

While several methods have been proposed to tackle this problem, there is limited work explaining why these methods work well. This paper has the goal of better explaining a popularly used technique for avoiding catastrophic forgetting: quadratic regularization. We show that quadratic regularizers prevent forgetting of past tasks by interpolating current and previous values of model parameters at every training iteration. Over multiple training iterations, this interpolation operation reduces the learning rates of more important model parameters, thereby minimizing their movement. Our analysis also reveals two drawbacks of quadratic regularization: (a) dependence of parameter interpolation on training hyperparameters, which often leads to training instability and (b) assignment of lower importance to deeper layers, which are generally the place forgetting occurs in DNNs. Via a simple modification to the order of operations, we show these drawbacks can be easily avoided, resulting in 6.2% higher average accuracy at 4.5% lower average forgetting. We confirm the robustness of our results by training over 2000 models in different settings. Learning algorithms are often designed under the assumption of independent and identical data distributions. However, this assumption is violated in several practical scenarios, such as continual learning and lifelong learning, where data distributions evolve constantly. In such settings, deep neural networks (DNNs) witness catastrophic forgetting and have difficulty adapting to new tasks without losing performance on previously learned ones. Several past works have tried to address this problem. While these works have shown promising results, a detailed understanding of their proposed methods is still to be developed. Understanding the reasons due to which existing methods for preventing catastrophic forgetting work or fail can open the possibility of developing better methods. With this motivation, in this work, we analyze quadratic regularization, a popular technique for preventing catastrophic forgetting in DNNs.