Loss of plasticity in deep neural networks is the gradual reduction in a model's capacity to incrementally learn and has been identified as a key obstacle to learning in non-stationary problem settings. Recent work has shown that deep linear networks tend to be resilient towards loss of plasticity. Motivated by this observation, we propose Adaptive Linearization (AdaLin), a general approach that dynamically adapts each neuron's activation function to mitigate plasticity loss. Unlike prior methods that rely on regularization or periodic resets, AdaLin equips every neuron with a learnable parameter and a gating mechanism that injects linearity into the activation function based on its gradient flow. This adaptive modulation ensures sufficient gradient signal and sustains continual learning without introducing additional hyperparameters or requiring explicit task boundaries. When used with conventional activation functions like ReLU, Tanh, and GeLU, we demonstrate that AdaLin can significantly improve performance on standard benchmarks, including Random Label and Permuted MNIST, Random Label and Shuffled CIFAR-10, and Class-Split CIFAR-100. Furthermore, its efficacy is shown in more complex scenarios, such as class-incremental learning on CIFAR-100 with a ResNet-18 backbone, and in mitigating plasticity loss in off-policy reinforcement learning agents. We perform a systematic set of ablations that show that neuron-level adaptation is crucial for good performance and analyze a number of metrics in the network that might be correlated to loss of plasticity.
Biological synapses effortlessly balance memory retention and flexibility, yet artificial neural networks still struggle with the extremes of catastrophic forgetting and catastrophic remembering. Here, we introduce Metaplasticity from Synaptic Uncertainty (MESU), a Bayesian framework that updates network parameters according their uncertainty. This approach allows a principled combination of learning and forgetting that ensures that critical knowledge is preserved while unused or outdated information is gradually released. Unlike standard Bayesian approaches -- which risk becoming overly constrained, and popular continual-learning methods that rely on explicit task boundaries, MESU seamlessly adapts to streaming data. It further provides reliable epistemic uncertainty estimates, allowing out-of-distribution detection, the only computational cost being to sample the weights multiple times to provide proper output statistics. Experiments on image-classification benchmarks demonstrate that MESU mitigates catastrophic forgetting, while maintaining plasticity for new tasks. When training 200 sequential permuted MNIST tasks, MESU outperforms established continual learning techniques in terms of accuracy, capability to learn additional tasks, and out-of-distribution data detection. Additionally, due to its non-reliance on task boundaries, MESU outperforms conventional learning techniques on the incremental training of CIFAR-100 tasks consistently in a wide range of scenarios. Our results unify ideas from metaplasticity, Bayesian inference, and Hessian-based regularization, offering a biologically-inspired pathway to robust, perpetual learning.
Continual learning (CL) has been an essential topic in the contemporary application of deep neural networks, where catastrophic forgetting (CF) can impede a model's ability to acquire knowledge progressively. Existing CL strategies primarily address CF by regularizing model updates or separating task-specific and shared components. However, these methods focus on task model elements while overlooking the potential of leveraging inter-task relationships for learning enhancement. To address this, we propose a transferability-aware task embedding named H-embedding and train a hypernet under its guidance to learn task-conditioned model weights for CL tasks. Particularly, H-embedding is introduced based on an information theoretical transferability measure and is designed to be online and easy to compute. The framework is also characterized by notable practicality, which only requires storing a low-dimensional task embedding for each task, and can be efficiently trained in an end-to-end way. Extensive evaluations and experimental analyses on datasets including Permuted MNIST, Cifar10/100, and ImageNet-R demonstrate that our framework performs prominently compared to various baseline methods, displaying great potential in exploiting intrinsic task relationships.
Continual learning of partially similar tasks poses a challenge for artificial neural networks, as task similarity presents both an opportunity for knowledge transfer and a risk of interference and catastrophic forgetting. However, it remains unclear how task similarity in input features and readout patterns influences knowledge transfer and forgetting, as well as how they interact with common algorithms for continual learning. Here, we develop a linear teacher-student model with latent structure and show analytically that high input feature similarity coupled with low readout similarity is catastrophic for both knowledge transfer and retention. Conversely, the opposite scenario is relatively benign. Our analysis further reveals that task-dependent activity gating improves knowledge retention at the expense of transfer, while task-dependent plasticity gating does not affect either retention or transfer performance at the over-parameterized limit. In contrast, weight regularization based on the Fisher information metric significantly improves retention, regardless of task similarity, without compromising transfer performance. Nevertheless, its diagonal approximation and regularization in the Euclidean space are much less robust against task similarity. We demonstrate consistent results in a permuted MNIST task with latent variables. Overall, this work provides insights into when continual learning is difficult and how to mitigate it.
Catastrophic forgetting remains a challenge for neural networks, especially in lifelong learning scenarios. In this study, we introduce MEtaplasticity from Synaptic Uncertainty (MESU), inspired by metaplasticity and Bayesian inference principles. MESU harnesses synaptic uncertainty to retain information over time, with its update rule closely approximating the diagonal Newton's method for synaptic updates. Through continual learning experiments on permuted MNIST tasks, we demonstrate MESU's remarkable capability to maintain learning performance across 100 tasks without the need of explicit task boundaries.
Biological neural networks are capable of recruiting different sets of neurons to encode different memories. However, when training artificial neural networks on a set of tasks, typically, no mechanism is employed for selectively producing anything analogous to these neuronal ensembles. Further, artificial neural networks suffer from catastrophic forgetting, where the network's performance rapidly deteriorates as tasks are learned sequentially. By contrast, sequential learning is possible for a range of biological organisms. We introduce Learned Context Dependent Gating (LXDG), a method to flexibly allocate and recall `artificial neuronal ensembles', using a particular network structure and a new set of regularization terms. Activities in the hidden layers of the network are modulated by gates, which are dynamically produced during training. The gates are outputs of networks themselves, trained with a sigmoid output activation. The regularization terms we have introduced correspond to properties exhibited by biological neuronal ensembles. The first term penalizes low gate sparsity, ensuring that only a specified fraction of the network is used. The second term ensures that previously learned gates are recalled when the network is presented with input from previously learned tasks. Finally, there is a regularization term responsible for ensuring that new tasks are encoded in gates that are as orthogonal as possible from previously used ones. We demonstrate the ability of this method to alleviate catastrophic forgetting on continual learning benchmarks. When the new regularization terms are included in the model along with Elastic Weight Consolidation (EWC) it achieves better performance on the benchmark `permuted MNIST' than with EWC alone. The benchmark `rotated MNIST' demonstrates how similar tasks recruit similar neurons to the artificial neuronal ensemble.
Overparameterization is known to permit strong generalization performance in neural networks. In this work, we provide an initial theoretical analysis of its effect on catastrophic forgetting in a continual learning setup. We show experimentally that in permuted MNIST image classification tasks, the generalization performance of multilayer perceptrons trained by vanilla stochastic gradient descent can be improved by overparameterization, and the extent of the performance increase achieved by overparameterization is comparable to that of state-of-the-art continual learning algorithms. We provide a theoretical explanation of this effect by studying a qualitatively similar two-task linear regression problem, where each task is related by a random orthogonal transformation. We show that when a model is trained on the two tasks in sequence without any additional regularization, the risk gain on the first task is small if the model is sufficiently overparameterized.
Purpose: We propose a novel method for continual learning based on the increasing depth of neural networks. This work explores whether extending neural network depth may be beneficial in a life-long learning setting. Methods: We propose a novel approach based on adding new layers on top of existing ones to enable the forward transfer of knowledge and adapting previously learned representations. We employ a method of determining the most similar tasks for selecting the best location in our network to add new nodes with trainable parameters. This approach allows for creating a tree-like model, where each node is a set of neural network parameters dedicated to a specific task. The Progressive Neural Network concept inspires the proposed method. Therefore, it benefits from dynamic changes in network structure. However, Progressive Neural Network allocates a lot of memory for the whole network structure during the learning process. The proposed method alleviates this by adding only part of a network for a new task and utilizing a subset of previously trained weights. At the same time, we may retain the benefit of PNN, such as no forgetting guaranteed by design, without needing a memory buffer. Results: Experiments on Split CIFAR and Split Tiny ImageNet show that the proposed algorithm is on par with other continual learning methods. In a more challenging setup with a single computer vision dataset as a separate task, our method outperforms Experience Replay. Conclusion: It is compatible with commonly used computer vision architectures and does not require a custom network structure. As an adaptation to changing data distribution is made by expanding the architecture, there is no need to utilize a rehearsal buffer. For this reason, our method could be used for sensitive applications where data privacy must be considered.
Continual learning (CL) is widely regarded as crucial challenge for lifelong AI. However, existing CL benchmarks, e.g. Permuted-MNIST and Split-CIFAR, make use of artificial temporal variation and do not align with or generalize to the real-world. In this paper, we introduce CLEAR, the first continual image classification benchmark dataset with a natural temporal evolution of visual concepts in the real world that spans a decade (2004-2014). We build CLEAR from existing large-scale image collections (YFCC100M) through a novel and scalable low-cost approach to visio-linguistic dataset curation. Our pipeline makes use of pretrained vision-language models (e.g. CLIP) to interactively build labeled datasets, which are further validated with crowd-sourcing to remove errors and even inappropriate images (hidden in original YFCC100M). The major strength of CLEAR over prior CL benchmarks is the smooth temporal evolution of visual concepts with real-world imagery, including both high-quality labeled data along with abundant unlabeled samples per time period for continual semi-supervised learning. We find that a simple unsupervised pre-training step can already boost state-of-the-art CL algorithms that only utilize fully-supervised data. Our analysis also reveals that mainstream CL evaluation protocols that train and test on iid data artificially inflate performance of CL system. To address this, we propose novel "streaming" protocols for CL that always test on the (near) future. Interestingly, streaming protocols (a) can simplify dataset curation since today's testset can be repurposed for tomorrow's trainset and (b) can produce more generalizable models with more accurate estimates of performance since all labeled data from each time-period is used for both training and testing (unlike classic iid train-test splits).
One weakness of machine learning algorithms is the poor ability of models to solve new problems without forgetting previously acquired knowledge. The Continual Learning (CL) paradigm has emerged as a protocol to systematically investigate settings where the model sequentially observes samples generated by a series of tasks. In this work, we take a task-agnostic view of continual learning and develop a hierarchical information-theoretic optimality principle that facilitates a trade-off between learning and forgetting. We discuss this principle from a Bayesian perspective and show its connections to previous approaches to CL. Based on this principle, we propose a neural network layer, called the Mixture-of-Variational-Experts layer, that alleviates forgetting by creating a set of information processing paths through the network which is governed by a gating policy. Due to the general formulation based on generic utility functions, we can apply this optimality principle to a large variety of learning problems, including supervised learning, reinforcement learning, and generative modeling. We demonstrate the competitive performance of our method in continual supervised learning and in continual reinforcement learning.
Continual learning tries to learn new tasks without forgetting previously learned ones. In reality, most of the existing artificial neural network(ANN) models fail, while humans do the same by remembering previous works throughout their life. Although simply storing all past data can alleviate the problem, it needs large memory and often infeasible in real-world applications where last data access is limited. We hypothesize that the model that learns to solve each task continually has some task-specific properties and some task-invariant characteristics. We propose a hybrid continual learning model that is more suitable in real case scenarios to address the issues that has a task-invariant shared variational autoencoder and T task-specific variational autoencoders. Our model combines generative replay and architectural growth to prevent catastrophic forgetting. We show our hybrid model effectively avoids forgetting and achieves state-of-the-art results on visual continual learning benchmarks such as MNIST, Permuted MNIST(QMNIST), CIFAR100, and miniImageNet datasets. We discuss results on a few more datasets, such as SVHN, Fashion-MNIST, EMNIST, and CIFAR10.
Preventing catastrophic forgetting while continually learning new tasks is an essential problem in lifelong learning. Structural regularization (SR) refers to a family of algorithms that mitigate catastrophic forgetting by penalizing the network for changing its "critical parameters" from previous tasks while learning a new one. The penalty is often induced via a quadratic regularizer defined by an \emph{importance matrix}, e.g., the (empirical) Fisher information matrix in the Elastic Weight Consolidation framework. In practice and due to computational constraints, most SR methods crudely approximate the importance matrix by its diagonal. In this paper, we propose \emph{Sketched Structural Regularization} (Sketched SR) as an alternative approach to compress the importance matrices used for regularizing in SR methods. Specifically, we apply \emph{linear sketching methods} to better approximate the importance matrices in SR algorithms. We show that sketched SR: (i) is computationally efficient and straightforward to implement, (ii) provides an approximation error that is justified in theory, and (iii) is method oblivious by construction and can be adapted to any method that belongs to the structural regularization class. We show that our proposed approach consistently improves various SR algorithms' performance on both synthetic experiments and benchmark continual learning tasks, including permuted-MNIST and CIFAR-100.
Conventional neural architectures for sequential data present important limitations. Recurrent networks suffer from exploding and vanishing gradients, small effective memory horizons, and must be trained sequentially. Convolutional networks are unable to handle sequences of unknown size and their memory horizon must be defined a priori. In this work, we show that all these problems can be solved by formulating convolutional kernels in CNNs as continuous functions. The resulting Continuous Kernel Convolution (CKConv) allows us to model arbitrarily long sequences in a parallel manner, within a single operation, and without relying on any form of recurrence. We show that Continuous Kernel Convolutional Networks (CKCNNs) obtain state-of-the-art results in multiple datasets, e.g., permuted MNIST, and, thanks to their continuous nature, are able to handle non-uniformly sampled datasets and irregularly-sampled data natively. CKCNNs match or perform better than neural ODEs designed for these purposes in a faster and simpler manner.
Training RNNs to learn long-term dependencies is difficult due to vanishing gradients. We explore an alternative solution based on explicit memorization using linear autoencoders for sequences, which allows to maximize the short-term memory and that can be solved with a closed-form solution without backpropagation. We introduce an initialization schema that pretrains the weights of a recurrent neural network to approximate the linear autoencoder of the input sequences and we show how such pretraining can better support solving hard classification tasks with long sequences. We test our approach on sequential and permuted MNIST. We show that the proposed approach achieves a much lower reconstruction error for long sequences and a better gradient propagation during the finetuning phase.
A central problem in learning from sequential data is representing cumulative history in an incremental fashion as more data is processed. We introduce a general framework (HiPPO) for the online compression of continuous signals and discrete time series by projection onto polynomial bases. Given a measure that specifies the importance of each time step in the past, HiPPO produces an optimal solution to a natural online function approximation problem. As special cases, our framework yields a short derivation of the recent Legendre Memory Unit (LMU) from first principles, and generalizes the ubiquitous gating mechanism of recurrent neural networks such as GRUs. This formal framework yields a new memory update mechanism (HiPPO-LegS) that scales through time to remember all history, avoiding priors on the timescale. HiPPO-LegS enjoys the theoretical benefits of timescale robustness, fast updates, and bounded gradients. By incorporating the memory dynamics into recurrent neural networks, HiPPO RNNs can empirically capture complex temporal dependencies. On the benchmark permuted MNIST dataset, HiPPO-LegS sets a new state-of-the-art accuracy of 98.3%. Finally, on a novel trajectory classification task testing robustness to out-of-distribution timescales and missing data, HiPPO-LegS outperforms RNN and neural ODE baselines by 25-40% accuracy.
Continual learning is the problem of sequentially learning new tasks or knowledge while protecting previously acquired knowledge. However, catastrophic forgetting poses a grand challenge for neural networks performing such learning process. Thus, neural networks that are deployed in the real world often struggle in scenarios where the data distribution is non-stationary (concept drift), imbalanced, or not always fully available, i.e., rare edge cases. We propose a Differentiable Hebbian Consolidation model which is composed of a Differentiable Hebbian Plasticity (DHP) Softmax layer that adds a rapid learning plastic component (compressed episodic memory) to the fixed (slow changing) parameters of the softmax output layer; enabling learned representations to be retained for a longer timescale. We demonstrate the flexibility of our method by integrating well-known task-specific synaptic consolidation methods to penalize changes in the slow weights that are important for each target task. We evaluate our approach on the Permuted MNIST, Split MNIST and Vision Datasets Mixture benchmarks, and introduce an imbalanced variant of Permuted MNIST -- a dataset that combines the challenges of class imbalance and concept drift. Our proposed model requires no additional hyperparameters and outperforms comparable baselines by reducing forgetting.
We propose a quadratic penalty method for continual learning of neural networks that contain batch normalization (BN) layers. The Hessian of a loss function represents the curvature of the quadratic penalty function, and a Kronecker-factored approximate curvature (K-FAC) is used widely to practically compute the Hessian of a neural network. However, the approximation is not valid if there is dependence between examples, typically caused by BN layers in deep network architectures. We extend the K-FAC method so that the inter-example relations are taken into account and the Hessian of deep neural networks can be properly approximated under practical assumptions. We also propose a method of weight merging and reparameterization to properly handle statistical parameters of BN, which plays a critical role for continual learning with BN, and a method that selects hyperparameters without source task data. Our method shows better performance than baselines in the permuted MNIST task with BN layers and in sequential learning from the ImageNet classification task to fine-grained classification tasks with ResNet-50, without any explicit or implicit use of source task data for hyperparameter selection.
In the continual learning setting, tasks are encountered sequentially. The goal is to learn whilst i) avoiding catastrophic forgetting, ii) efficiently using model capacity, and iii) employing forward and backward transfer learning. In this paper, we explore how the Variational Continual Learning (VCL) framework achieves these desiderata on two benchmarks in continual learning: split MNIST and permuted MNIST. We first report significantly improved results on what was already a competitive approach. The improvements are achieved by establishing a new best practice approach to mean-field variational Bayesian neural networks. We then look at the solutions in detail. This allows us to obtain an understanding of why VCL performs as it does, and we compare the solution to what an `ideal' continual learning solution might be.
Standard artificial neural networks suffer from the well-known issue of catastrophic forgetting, making continual or lifelong learning difficult for machine learning. In recent years, numerous methods have been proposed for continual learning, but due to differences in evaluation protocols it is difficult to directly compare their performance. To enable more structured comparisons, we describe three continual learning scenarios based on whether at test time task identity is provided and--in case it is not--whether it must be inferred. Any sequence of well-defined tasks can be performed according to each scenario. Using the split and permuted MNIST task protocols, for each scenario we carry out an extensive comparison of recently proposed continual learning methods. We demonstrate substantial differences between the three scenarios in terms of difficulty and in terms of how efficient different methods are. In particular, when task identity must be inferred (i.e., class incremental learning), we find that regularization-based approaches (e.g., elastic weight consolidation) fail and that replaying representations of previous experiences seems required for solving this scenario.
Addressing catastrophic forgetting is one of the key challenges in continual learning where machine learning systems are trained with sequential or streaming tasks. Despite recent remarkable progress in state-of-the-art deep learning, deep neural networks (DNNs) are still plagued with the catastrophic forgetting problem. This paper presents a conceptually simple yet general and effective framework for handling catastrophic forgetting in continual learning with DNNs. The proposed method consists of two components: a neural structure optimization component and a parameter learning and/or fine-tuning component. By separating the explicit neural structure learning and the parameter estimation, not only is the proposed method capable of evolving neural structures in an intuitively meaningful way, but also shows strong capabilities of alleviating catastrophic forgetting in experiments. Furthermore, the proposed method outperforms all other baselines on the permuted MNIST dataset, the split CIFAR100 dataset and the Visual Domain Decathlon dataset in continual learning setting.