Research: Learning Algorithms and Optimization

Learning Algorithms and Optimization

(Return to Current Research Outline)

B. Continual Learning and Unlearning

Humans have the remarkable ability to learn new topics without relearning everything from scratch. While continual learning algorithms have been explored extensively in research, they often suffer from inefficiencies in memory complexity and training time. Our lab develops algorithms that address these bottlenecks by examining the mathematics of learning algorithms. Furthermore, we are pioneers in proposing unlearning algorithms, inspired by the human ability to "unlearn" or correct previously acquired biases without starting anew. In an era where data privacy and model efficiency are critical, unlearning algorithms play a pivotal role in enhancing AI systems.

Publications (Continual Learning):

1. Awasthi, Ankita, Marco Apolinario, and Kaushik Roy. "MANGO: Meta-Adaptive Network Gradient Optimization for Online Continual Learning." ArXiv, 2026.

   Abstract: In Online Continual Learning (OCL), a neural network sequentially learns from a non-stationary data stream in a single-pass with access only to a limited memory replay buffer. This contrasts sharply with off-line continual learning where training is multiple epoch dependent on large datasets. The main challenge faced by OCL is to overcome catastrophic forgetting of past tasks (stability) while learning new ones efficiently (plasticity). Existing methods counter forgetting via replay-based rehearsal, output level distillation, fixed regularization, or meta-learning on the current data. However, these methods have limitations: rehearsal introduces a stored sample bias; distillation operates on output-distributions without modulating parameter updates; fixed-regularization penalizes parameters irrespective of sensitivity; stream-only meta-learning lacks a feedback controlled parameter update. We propose Meta-Adaptive Network Gradient Optimization (MANGO), an OCL framework that balances stability-plasticity via gradient-gating and meta-learned regularization. Gradient-gating scales parameter updates based on sensitivity, preventing destructive updates. Meta-learned regularization adapts stability coefficients, evaluating the effect of parameter update on replay. In MANGO, replay acts as both a training signal and a forgetting evaluator. We evaluated our method on three standard OCL benchmark datasets. MANGO outperforms strong baselines, achieving state-of-the-art results with consistent performance across replay sizes. In domain incremental learning on CLEAR-10 and class incremental learning on CIFAR-100 and Tiny-ImageNet, it achieves highest accuracy among all baselines and achieves positive Backward Transfer, overcoming forgetting on CLEAR-10.

2. Apolinario, Marco Paul E and Kaushik Roy. "LANCE: Low Rank Activation Compression for Efficient On-Device Continual Learning." European Conference on Computer Vision (ECCV), 2026.

   Abstract: On-device learning is essential for personalization, privacy, and long-term adaptation in resource-constrained environments. Achieving this requires efficient learning, both fine-tuning existing models and continually acquiring new tasks without catastrophic forgetting. Yet both settings are constrained by high memory cost of storing activations during backpropagation. Existing activation compression methods reduce this cost but rely on repeated low-rank decompositions, introducing computational overhead. Also, such methods have not been explored for continual learning. We propose LANCE (Low-rank Activation Compression), a framework that performs one-shot higher-order Singular Value Decomposition (SVD) to obtain a reusable low-rank subspace for activation projection. This eliminates repeated decompositions, reducing both memory and computation. Moreover, fixed low-rank subspaces further enable on-device continual learning by allocating tasks to orthogonal subspaces without storing large task-specific matrices. Experiments show that LANCE reduces activation storage up to 250× while maintaining accuracy comparable to full backpropagation on CIFAR-10/100, Oxford-IIIT Pets, Flowers102, and CUB-200 datasets. On continual learning benchmarks (Split CIFAR-100, Split MiniImageNet, 5-Datasets), it performs competitively with orthogonal gradient projection methods at a fraction of the memory cost. These results position LANCE as a practical and scalable solution for efficient fine-tuning and continual learning on edge devices.

3. Biswas, Shristi Das, Yue Zhang, Anwesan Pal, Radhika Bhargava, Kaushik Roy. "ELLA: Efficient Lifelong Learning for Adapters in Large Language Models." European Chapter of the Association for Computational Linguistics (EACL), 2026.

   Abstract: Large Language Models (LLMs) suffer severe catastrophic forgetting when adapted sequentially to new tasks in a continual learning (CL) setting. Existing approaches are fundamentally limited: replay-based methods are impractical and privacy-violating, while strict orthogonality-based methods collapse under scale: each new task is projected onto an orthogonal complement, progressively reducing the residual degrees of freedom and eliminating forward transfer by forbidding overlap in shared representations. In this work, we introduce ELLA, a training framework built on the principle of selective subspace de-correlation. Rather than forbidding all overlap, ELLA explicitly characterizes the structure of past updates and penalizes alignments along their high-energy, task-specific directions, while preserving freedom in the low-energy residual subspaces to enable transfer. Formally, this is realized via a lightweight regularizer on a single aggregated update matrix. We prove this mechanism corresponds to an anisotropic shrinkage operator that bounds interference, yielding a penalty that is both memory- and compute-constant regardless of task sequence length. ELLA requires no data replay, no architectural expansion, and negligible storage. Empirically, it achieves state-of-the-art CL performance on three popular benchmarks, with relative accuracy gains of up to and a smaller memory footprint. Further, ELLA scales robustly across architectures and actively enhances the model's zero-shot generalization performance on unseen tasks, establishing a principled and scalable solution for constructive lifelong LLM adaptation.

4. Apolinario, Marco PE, Sakshi Choudhary, Kaushik Roy. "CODE-CL: Conceptor-Based Gradient Projection for Deep Continual Learning." IEEE/CVF International Conference on Computer Vision (WACV), 2025.

   Abstract: Continual learning (CL) -- the ability to progressively acquire and integrate new concepts -- is essential to intelligent systems to adapt to dynamic environments. However, deep neural networks struggle with catastrophic forgetting (CF) when learning tasks sequentially, as training for new tasks often overwrites previously learned knowledge. To address this, recent approaches constrain updates to orthogonal subspaces using gradient projection, effectively preserving important gradient directions for previous tasks. While effective in reducing forgetting, these approaches inadvertently hinder forward knowledge transfer (FWT), particularly when tasks are highly correlated. In this work, we propose Conceptor-based gradient projection for Deep Continual Learning (CODE-CL), a novel method that leverages conceptor matrix representations, a form of regularized reconstruction, to adaptively handle highly correlated tasks. CODE-CL mitigates CF by projecting gradients onto pseudo-orthogonal subspaces of previous task feature spaces while simultaneously promoting FWT. It achieves this by learning a linear combination of shared basis directions, allowing efficient balance between stability and plasticity and transfer of knowledge between overlapping input feature representations. Extensive experiments on continual learning benchmarks validate CODE-CL's efficacy, demonstrating superior performance, reduced forgetting, and improved FWT as compared to state-of-the-art methods.

5. Saha, Gobinda, and Kaushik Roy. "Online continual learning with saliency-guided experience replay using tiny episodic memory." Machine Vision and Applications 34.4 (2023): 65. CODE AVAILABLE HERE

   Abstract: Artificial learning systems aspire to mimic human intelligence by continually learning from a stream of tasks without forgetting past knowledge. One way to enable such learning is to store past experiences in the form of input examples in episodic memory and replay them when learning new tasks. However, performance of such method suffers as the size of the memory becomes smaller. In this paper, we propose a new approach for experience replay, where we select the past experiences by looking at the saliency maps, which provide visual explanations for the model?s decision. Guided by these saliency maps, we pack the memory with only the parts or patches of the input images important for the model?s prediction. While learning a new task, we replay these memory patches with appropriate zero-padding to remind the model about its past decisions. We evaluate our algorithm on CIFAR-100, miniImageNet and CUB datasets and report better performance than the state-of-the-art approaches. We perform a detailed study to show the effectiveness of zero-padded patch replay compared to the other candidate approaches. Moreover, with qualitative and quantitative analyses we show that our method captures richer summaries of past experiences without any memory increase and hence performs well with small episodic memory.

6. Saha, Gobinda, and Kaushik Roy. "Continual learning with scaled gradient projection." Proceedings of the AAAI Conference on Artificial Intelligence. Vol. 37. No. 8. 2023. CODE AVAILABLE HERE

   Abstract: In neural networks, continual learning results in gradient interference among sequential tasks, leading to catastrophic forgetting of old tasks while learning new ones. This issue is addressed in recent methods by storing the important gradient spaces for old tasks and updating the model orthogonally during new tasks. However, such restrictive orthogonal gradient updates hamper the learning capability of the new tasks resulting in sub-optimal performance. To improve new learning while minimizing forgetting, in this paper we propose a Scaled Gradient Projection (SGP) method, where we combine the orthogonal gradient projections with scaled gradient steps along the important gradient spaces for the past tasks. The degree of gradient scaling along these spaces depends on the importance of the bases spanning them. We propose an efficient method for computing and accumulating importance of these bases using the singular value decomposition of the input representations for each task. We conduct extensive experiments ranging from continual image classification to reinforcement learning tasks and report better performance with less training overhead than the state-of-the-art approaches.

7. Saha, Gobinda, Isha Garg, and Kaushik Roy. "Gradient Projection Memory for Continual Learning." International Conference on Learning Representations, (2021). CODE AVAILABLE HERE

   Abstract: The ability to learn continually without forgetting the past tasks is a desired attribute for artificial learning systems. Existing approaches to enable such learning in artificial neural networks usually rely on network growth, importance-based weight update or replay of old data from the memory. In contrast, we propose a novel approach where a neural network learns new tasks by taking gradient steps in the orthogonal direction to the gradient subspaces deemed important for the past tasks. We find the bases of these subspaces by analyzing network representations (activations) after learning each task with Singular Value Decomposition (SVD) in a single shot manner and store them in the memory as Gradient Projection Memory (GPM). With qualitative and quantitative analyses, we show that such orthogonal gradient descent induces minimum to no interference with the past tasks, thereby mitigates forgetting. We evaluate our algorithm on diverse image classification datasets with short and long sequences of tasks and report better or on-par performance compared to the state-of-the-art approaches.

8. Saha, Gobinda, Isha Garg, Aayush Ankit, and Kaushik Roy. "Space: Structured compression and sharing of representational space for continual learning." IEEE Access 9 (2021): 150480-150494. CODE AVAILABLE HERE

   Abstract: Humans learn incrementally from sequential experiences throughout their lives, which has proven hard to emulate in artificial neural networks. Incrementally learning tasks causes neural networks to overwrite relevant information learned about older tasks, resulting in ?Catastrophic Forgetting?. Efforts to overcome this phenomenon often utilize resources poorly, for instance, by growing the network architecture or needing to save parametric importance scores, or violate data privacy between tasks. To tackle this, we propose SPACE, an algorithm that enables a network to learn continually and efficiently by partitioning the learnt space into a Core space, that serves as the condensed knowledge base over previously learned tasks, and a Residual space, which is akin to a scratch space for learning the current task. After learning each task, the Residual is analyzed for redundancy, both within itself and with the learnt Core space. A minimal number of extra dimensions required to explain the current task are added to the Core space and the remaining Residual is freed up for learning the next task. We evaluate our algorithm on P-MNIST, CIFAR and a sequence of 8 different datasets, and achieve comparable accuracy to the state-of-the-art methods while overcoming catastrophic forgetting. Additionally, our algorithm is well suited for practical use. The partitioning algorithm analyzes all layers in one shot, ensuring scalability to deeper networks. Moreover, the analysis of dimensions translates to filter-level sparsity, and the structured nature of the resulting architecture gives us up to 5x improvement in energy efficiency during task inference over the current state-of-the-art.

Publications (Unlearning):

1. Biswas, Shristi Das, Arani Roy, Kaushik Roy. "Now You See It, Now You Don't: Instant Concept Erasure for Safe Text-to-Image and Video Generation." IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR Findings), 2026.

   Abstract: Robust concept removal for text-to-image (T2I) and text-to-video (T2V) models is essential for their safe deployment. Existing methods, however, suffer from costly retraining, inference overhead, or vulnerability to adversarial attacks. Crucially, they rarely model the latent semantic overlap between the target erase concept and surrounding content -- causing collateral damage post-erasure -- and even fewer methods work reliably across both T2I and T2V domains. We introduce Instant Concept Erasure (ICE), a training-free, modality-agnostic, one-shot weight modification approach that achieves precise, persistent unlearning with zero overhead. ICE defines erase and preserve subspaces using anisotropic energy-weighted scaling, then explicitly regularizes against their intersection using a unique, closed-form overlap projector. We pose a convex and Lipschitz-bounded Spectral Unlearning Objective, balancing erasure fidelity and intersection preservation, that admits a stable and unique analytical solution. This solution defines a dissociation operator that is translated to the model's text-conditioning layers, making the edit permanent and runtime-free. Across targeted removals of artistic styles, objects, identities, and explicit content, ICE efficiently achieves strong erasure with improved robustness to red-teaming, all while causing only minimal degradation of original generative abilities in both T2I and T2V models.

2. Biswas, Shristi Das*, Arani Roy*, Kaushik Roy. "CURE: Concept Unlearning via Orthogonal Representation Editing in Diffusion Models." Advances in Neural Information Processing Systems (NeurIPS), 2026.

   Abstract: As Text-to-Image models continue to evolve, so does the risk of generating unsafe, copyrighted, or privacy-violating content. Existing safety interventions - ranging from training data curation and model fine-tuning to inference-time filtering and guidance - often suffer from incomplete concept removal, susceptibility to jail-breaking, computational inefficiency, or collateral damage to unrelated capabilities. In this paper, we introduce CURE, a training-free concept unlearning framework that operates directly in the weight space of pre-trained diffusion models, enabling fast, interpretable, and highly specific suppression of undesired concepts. At the core of our method is the Spectral Eraser, a closed-form, orthogonal projection module that identifies discriminative subspaces using Singular Value Decomposition over token embeddings associated with the concepts to forget and retain. Intuitively, the Spectral Eraser identifies and isolates features unique to the undesired concept while preserving safe attributes. This operator is then applied in a single step update to yield an edited model in which the target concept is effectively unlearned - without retraining, supervision, or iterative optimization. To balance the trade-off between filtering toxicity and preserving unrelated concepts, we further introduce an Expansion Mechanism for spectral regularization which selectively modulates singular vectors based on their relative significance to control the strength of forgetting. All the processes above are in closed-form, guaranteeing extremely efficient erasure in only 2 seconds. Benchmarking against prior approaches, CURE achieves a more efficient and thorough removal for targeted artistic styles, objects, identities, or explicit content, with minor damage to original generation ability and demonstrates enhanced robustness against red-teaming. Project Page at \url{https://sites.google.com/view/cure-unlearning/home}.

3. Kodge, Sangamesh, Deepak Ravikumar, Gobinda Saha, Kaushik Roy. "SAP: Corrective Machine Unlearning with Scaled Activation Projection for Label Noise Robustness." AAAI Conference on Artificial Intelligence (AAAI), 2025.

   Abstract: Label corruption, where training samples are mislabeled due to non-expert annotation or adversarial attacks, significantly degrades model performance. Acquiring large, perfectly labeled datasets is costly, and retraining models from scratch is computationally expensive. To address this, we introduce Scaled Activation Projection (SAP), a novel SVD (Singular Value Decomposition)-based corrective machine unlearning algorithm. SAP mitigates label noise by identifying a small subset of trusted samples using cross-entropy loss and projecting model weights onto a clean activation space estimated using SVD on these trusted samples. This process suppresses the noise introduced in activations due to the mislabeled samples. In our experiments, we demonstrate SAP?s effectiveness on synthetic noise with different settings and real-world label noise. SAP applied to the CIFAR dataset with 25% synthetic corruption show upto 6% generalization improvements. Additionally, SAP can improve the generalization over noise robust training approaches on CIFAR dataset by ? 3.2% on average. Further, we observe generalization improvements of 2.31% for a Vision Transformer model trained on naturally corrupted Clothing1M.

4. Kodge, Sangamesh, Gobinda Saha, and Kaushik Roy. "Deep Unlearning: Fast and Efficient Gradient-free Approach to Class Forgetting." Transactions on Machine Learning Research (TMLR), 2024. CODE AVAILABLE HERE

   Abstract: Machine unlearning is a prominent and challenging field, driven by regulatory demands for user data deletion and heightened privacy awareness. Existing approaches involve retraining model or multiple finetuning steps for each deletion request, often constrained by computational limits and restricted data access. In this work, we introduce a novel class unlearning algorithm designed to strategically eliminate specific classes from the learned model. Our algorithm first estimates the Retain and the Forget Spaces using Singular Value Decomposition on the layerwise activations for a small subset of samples from the retain and unlearn classes, respectively. We then compute the shared information between these spaces and remove it from the forget space to isolate class-discriminatory feature space. Finally, we obtain the unlearned model by updating the weights to suppress the class discriminatory features from the activation spaces. We demonstrate our algorithm's efficacy on ImageNet using a Vision Transformer with only ?1.5% drop in retain accuracy compared to the original model while maintaining under 1% accuracy on the unlearned class samples. Furthermore, our algorithm exhibits competitive unlearning performance and resilience against Membership Inference Attacks (MIA). Compared to baselines, it achieves an average accuracy improvement of 1.38% on the ImageNet dataset while requiring up to 10× fewer samples for unlearning. Additionally, under stronger MIA attacks on the CIFAR-100 dataset using a ResNet18 architecture, our approach outperforms the best baseline by 1.8%.

Click on the expansion arrows in each section to read the publication abstracts.

---