Closed-form merging of parameter-efficient modules for Federated Continual Learning

• URL: http://arxiv.org/abs/2410.17961v1
• Date: Wed, 23 Oct 2024 15:30:13 GMT
• Title: Closed-form merging of parameter-efficient modules for Federated Continual Learning
• Authors: Riccardo Salami, Pietro Buzzega, Matteo Mosconi, Jacopo Bonato, Luigi Sabetta, Simone Calderara,
• Abstract summary: We introduce LoRM, an alternating optimization strategy that trains one LoRA matrix at a time. This allows solving for each unknown variable individually, thus finding a unique solution. Our method demonstrates state-of-the-art performance across a range of FCIL scenarios.
• Score: 9.940242741914748
• License:
• Abstract: Model merging has emerged as a crucial technique in Deep Learning, enabling the integration of multiple models into a unified system while preserving performance and scalability. In this respect, the compositional properties of low-rank adaptation techniques (e.g., LoRA) have proven beneficial, as simple averaging LoRA modules yields a single model that mostly integrates the capabilities of all individual modules. Building on LoRA, we take a step further by imposing that the merged model matches the responses of all learned modules. Solving this objective in closed form yields an indeterminate system with A and B as unknown variables, indicating the existence of infinitely many closed-form solutions. To address this challenge, we introduce LoRM, an alternating optimization strategy that trains one LoRA matrix at a time. This allows solving for each unknown variable individually, thus finding a unique solution. We apply our proposed methodology to Federated Class-Incremental Learning (FCIL), ensuring alignment of model responses both between clients and across tasks. Our method demonstrates state-of-the-art performance across a range of FCIL scenarios.

Related papers

• Modeling Multi-Task Model Merging as Adaptive Projective Gradient Descent [74.02034188307857]
  Merging multiple expert models offers a promising approach for performing multi-task learning without accessing their original data. We find existing methods inevitably discard task-specific information that, while causing conflicts, is crucial for performance. Our approach consistently outperforms previous methods, achieving state-of-the-art results across diverse architectures and tasks in both vision and NLP domains.
  arXiv  Detail & Related papers  (2025-01-02T12:45:21Z)
• MAGIC++: Efficient and Resilient Modality-Agnostic Semantic Segmentation via Hierarchical Modality Selection [20.584588303521496]
  We introduce the MAGIC++ framework, which comprises two key plug-and-play modules for effective multi-modal fusion and hierarchical modality selection. Our method achieves state-of-the-art performance on both real-world and synthetic benchmarks. Our method is superior in the novel modality-agnostic setting, where it outperforms prior arts by a large margin.
  arXiv  Detail & Related papers  (2024-12-22T06:12:03Z)
• Unlocking Tuning-Free Few-Shot Adaptability in Visual Foundation Models by Recycling Pre-Tuned LoRAs [76.40876036912537]
  Large Language Models (LLMs) demonstrate strong few-shot adaptability without requiring fine-tuning. Current Visual Foundation Models (VFMs) require explicit fine-tuning with sufficient tuning data. We propose a framework, LoRA Recycle, that distills a meta-LoRA from diverse pre-tuned LoRAs with a meta-learning objective.
  arXiv  Detail & Related papers  (2024-12-03T07:25:30Z)
• Efficient and Effective Weight-Ensembling Mixture of Experts for Multi-Task Model Merging [111.8456671452411]
  Multi-task learning (MTL) leverages a shared model to accomplish multiple tasks and facilitate knowledge transfer. We propose a Weight-Ensembling Mixture of Experts (WEMoE) method for multi-task model merging. We show that WEMoE and E-WEMoE outperform state-of-the-art (SOTA) model merging methods in terms of MTL performance, generalization, and robustness.
  arXiv  Detail & Related papers  (2024-10-29T07:16:31Z)
• Federated Automatic Latent Variable Selection in Multi-output Gaussian Processes [0.7366405857677227]
  A common approach in MGPs to transfer knowledge across units involves gathering all data from each unit to a central server. We propose a hierarchical model that places spike-and-slab priors on the coefficients of each latent process. These priors help automatically select only needed latent processes by shrinking the coefficients of unnecessary ones to zero.
  arXiv  Detail & Related papers  (2024-07-24T02:03:28Z)
• LoRA-Ensemble: Efficient Uncertainty Modelling for Self-attention Networks [52.46420522934253]
  We introduce LoRA-Ensemble, a parameter-efficient deep ensemble method for self-attention networks. By employing a single pre-trained self-attention network with weights shared across all members, we train member-specific low-rank matrices for the attention projections. Our method exhibits superior calibration compared to explicit ensembles and achieves similar or better accuracy across various prediction tasks and datasets.
  arXiv  Detail & Related papers  (2024-05-23T11:10:32Z)
• FedSDD: Scalable and Diversity-enhanced Distillation for Model Aggregation in Federated Learning [15.39242780506777]
  We propose a scalable and diversity-enhanced federated distillation scheme, FedSDD, for federated learning. FedSDD decouples the training complexity from the number of clients to enhance the scalability, and builds the ensemble from a set of aggregated models. Experiment results show that FedSDD outperforms other FL methods, including FedAvg and FedDF, on the benchmark datasets.
  arXiv  Detail & Related papers  (2023-12-28T14:10:00Z)
• Cross-Silo Federated Learning Across Divergent Domains with Iterative Parameter Alignment [4.95475852994362]
  Federated learning is a method for training a machine learning model across remote clients. We reformulate the typical federated learning setup to learn N models optimized for a common objective. We find that the technique achieves competitive results on a variety of data partitions compared to state-of-the-art approaches.
  arXiv  Detail & Related papers  (2023-11-08T16:42:14Z)
• Module-wise Adaptive Distillation for Multimodality Foundation Models [125.42414892566843]
  multimodal foundation models have demonstrated remarkable generalizability but pose challenges for deployment due to their large sizes. One effective approach to reducing their sizes is layerwise distillation, wherein small student models are trained to match the hidden representations of large teacher models at each layer. Motivated by our observation that certain architecture components, referred to as modules, contribute more significantly to the student's performance than others, we propose to track the contributions of individual modules by recording the loss decrement after distillation each module and choose the module with a greater contribution to distill more frequently.
  arXiv  Detail & Related papers  (2023-10-06T19:24:00Z)
• Modular Deep Learning [120.36599591042908]
  Transfer learning has recently become the dominant paradigm of machine learning. It remains unclear how to develop models that specialise towards multiple tasks without incurring negative interference. Modular deep learning has emerged as a promising solution to these challenges.
  arXiv  Detail & Related papers  (2023-02-22T18:11:25Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.