Giulia Lanzillotta

Giulia Lanzillotta, Sidak Pal Singh, Benjamin F. Grewe et al. · eth-zurich

Continual learning is the problem of integrating new information in a model while retaining the knowledge acquired in the past. Despite the tangible improvements achieved in recent years, the problem of continual learning is still an open one. A better understanding of the mechanisms behind the successes and failures of existing continual learning algorithms can unlock the development of new successful strategies. In this work, we view continual learning from the perspective of the multi-task loss approximation, and we compare two alternative strategies, namely local and global approximations. We classify existing continual learning algorithms based on the approximation used, and we assess the practical effects of this distinction in common continual learning settings.Additionally, we study optimal continual learning objectives in the case of local polynomial approximations and we provide examples of existing algorithms implementing the optimal objectives

Giulia Lanzillotta, Sidak Pal Singh, Benjamin F. Grewe et al. · eth-zurich

Deep learning has proved to be a successful paradigm for solving many challenges in machine learning. However, deep neural networks fail when trained sequentially on multiple tasks, a shortcoming known as catastrophic forgetting in the continual learning literature. Despite a recent flourish of learning algorithms successfully addressing this problem, we find that provable guarantees against catastrophic forgetting are lacking. In this work, we study the relationship between learning and forgetting by looking at the geometry of neural networks' loss landscape. We offer a unifying perspective on a family of continual learning algorithms, namely methods based on parameter isolation, and we establish guarantees on catastrophic forgetting for some of them.

Giulia Lanzillotta, Damiano Meier, Thomas Hofmann

A persistent paradox in continual learning (CL) is that neural networks often retain linearly separable representations of past tasks even when their output predictions fail. We formalize this distinction as the gap between deep feature-space and shallow classifier-level forgetting. We reveal a critical asymmetry in Experience Replay: while minimal buffers successfully anchor feature geometry and prevent deep forgetting, mitigating shallow forgetting typically requires substantially larger buffer capacities. To explain this, we extend the Neural Collapse framework to the sequential setting. We characterize deep forgetting as a geometric drift toward out-of-distribution subspaces and prove that any non-zero replay fraction asymptotically guarantees the retention of linear separability. Conversely, we identify that the "strong collapse" induced by small buffers leads to rank-deficient covariances and inflated class means, effectively blinding the classifier to true population boundaries. By unifying CL with out-of-distribution detection, our work challenges the prevailing reliance on large buffers, suggesting that explicitly correcting these statistical artifacts could unlock robust performance with minimal replay.

Jacopo Graldi, Alessandro Breccia, Giulia Lanzillotta et al.

Despite recent efforts, neural networks still struggle to learn in non-stationary environments, and our understanding of catastrophic forgetting (CF) is far from complete. In this work, we perform a systematic study on the impact of model scale and the degree of feature learning in continual learning. We reconcile existing contradictory observations on scale in the literature, by differentiating between lazy and rich training regimes through a variable parameterization of the architecture. We show that increasing model width is only beneficial when it reduces the amount of feature learning, yielding more laziness. Using the framework of dynamical mean field theory, we then study the infinite width dynamics of the model in the feature learning regime and characterize CF, extending prior theoretical results limited to the lazy regime. We study the intricate relationship between feature learning, task non-stationarity, and forgetting, finding that high feature learning is only beneficial with highly similar tasks. We identify a transition modulated by task similarity where the model exits an effectively lazy regime with low forgetting to enter a rich regime with significant forgetting. Finally, our findings reveal that neural networks achieve optimal performance at a critical level of feature learning, which depends on task non-stationarity and transfers across model scales. This work provides a unified perspective on the role of scale and feature learning in continual learning.

Giulia Lanzillotta, Felix Sarnthein, Gil Kur et al.

The concept of knowledge distillation (KD) describes the training of a student model from a teacher model and is a widely adopted technique in deep learning. However, it is still not clear how and why distillation works. Previous studies focus on two central aspects of distillation: model size, and generalisation. In this work we study distillation in a third dimension: dataset size. We present a suite of experiments across a wide range of datasets, tasks and neural architectures, demonstrating that the effect of distillation is not only preserved but amplified in low-data regimes. We call this newly discovered property the data efficiency of distillation. Equipped with this new perspective, we test the predictive power of existing theories of KD as we vary the dataset size. Our results disprove the hypothesis that distillation can be understood as label smoothing, and provide further evidence in support of the dark knowledge hypothesis. Finally, we analyse the impact of modelling factors such as the objective, scale and relative number of samples on the observed phenomenon. Ultimately, this work reveals that the dataset size may be a fundamental but overlooked variable in the mechanisms underpinning distillation.

Amir Joudaki, Giulia Lanzillotta, Mohammad Samragh Razlighi et al. · utoronto, uw

Deep learning models excel in stationary data but struggle in non-stationary environments due to a phenomenon known as loss of plasticity (LoP), the degradation of their ability to learn in the future. This work presents a first-principles investigation of LoP in gradient-based learning. Grounded in dynamical systems theory, we formally define LoP by identifying stable manifolds in the parameter space that trap gradient trajectories. Our analysis reveals two primary mechanisms that create these traps: frozen units from activation saturation and cloned-unit manifolds from representational redundancy. Our framework uncovers a fundamental tension: properties that promote generalization in static settings, such as low-rank representations and simplicity biases, directly contribute to LoP in continual learning scenarios. We validate our theoretical analysis with numerical simulations and explore architectural choices or targeted perturbations as potential mitigation strategies.

Yuzhi Liu, Zixuan Chen, Zirui Zhang et al.

The Neural Tangent Kernel (NTK) offers a powerful tool to study the functional dynamics of neural networks. In the so-called lazy, or kernel regime, the NTK remains static during training and the network function is linear in the static neural tangents feature space. The evolution of the NTK during training is necessary for feature learning, a key driver of deep learning success. The study of the NTK dynamics has led to several critical discoveries in recent years, in generalization and scaling behaviours. However, this body of work has been limited to the single task setting, where the data distribution is assumed constant over time. In this work, we present a comprehensive empirical analysis of NTK dynamics in continual learning, where the data distribution shifts over time. Our findings highlight continual learning as a rich and underutilized testbed for probing the dynamics of neural training. At the same time, they challenge the validity of static-kernel approximations in theoretical treatments of continual learning, even at large scale.