Model merging is an efficient empowerment technique in the machine learning community that does not require the collection of raw training data and does not require expensive computation. As model merging becomes increasingly prevalent across various fields, it is crucial to understand the available model merging techniques comprehensively. However, there is a significant gap in the literature regarding a systematic and thorough review of these techniques. This survey provides a comprehensive overview of model merging methods and theories, their applications in various domains and settings, and future research directions. Specifically, we first propose a new taxonomic approach that exhaustively discusses existing model merging methods. Secondly, we discuss the application of model merging techniques in large language models, multimodal large language models, and more than ten machine learning subfields, including continual learning, multi-task learning, few-shot learning, and so on. Finally, we highlight the remaining challenges of model merging and discuss future research directions. A comprehensive list of papers about model merging is available at https://github.com/EnnengYang/Awesome-Model-Merging-Methods-Theories-Applications .
26 June 2023 in Proceedings of the AAAI Conference on Artificial Intelligence
Multi-task learning (MTL) models have demonstrated impressive results in computer vision, natural language processing, and recommender systems. Even though many approaches have been proposed, how well these approaches balance different tasks on each parameter still remains unclear. In this paper, we propose to measure the task dominance degree of a parameter by the total updates of each task on this parameter. Specifically, we compute the total updates by the exponentially decaying Average of the squared Updates (AU) on a parameter from the corresponding task. Based on this novel metric, we observe that many parameters in existing MTL methods, especially those in the higher shared layers, are still dominated by one or several tasks. The dominance of AU is mainly due to the dominance of accumulative gradients from one or several tasks. Motivated by this, we propose a Task-wise Adaptive learning rate approach, AdaTask in short, to separate the accumulative gradients and hence the learning rate of each task for each parameter in adaptive learning rate approaches (e.g., AdaGrad, RMSProp, and Adam). Comprehensive experiments on computer vision and recommender system MTL datasets demonstrate that AdaTask significantly improves the performance of dominated tasks, resulting SOTA average task-wise performance. Analysis on both synthetic and real-world datasets shows AdaTask balance parameters in every shared layer well.
13 October 2023 in IEEE Transactions on Knowledge and Data Engineering
Sequential recommendation has now been more widely studied, characterized by its well-consistency with real-world recommendation situations. Most existing works model user preference as the transition pattern from the previous item to the next, ignoring the time interval between these two items. However, we find that the time intervals in different sequences may vary significantly and thus result in the ineffectiveness of user modeling due to the issue of preference drift . Thus we propose an assumption that a sequence with uniformly distributed time intervals (denoted as uniform sequence) is more beneficial for preference learning than that with greatly varying time intervals. We then conduct an empirical study on four real datasets and the results support this assumption. Therefore, we advocate to augment sequence data from the perspective of time intervals, which is not studied in the literature. Specifically, we design five operators (Ti-Crop, Ti-CateReorder, Ti-Mask, Ti-Substitute, Ti-Insert) to transform the original non-uniform sequence to uniform sequence with the consideration of time intervals. Then, we devise a control strategy to execute data augmentation on item sequences in different lengths and a looseness range to ensure the generalization (or diversity) of generated data. Finally, we implement these improvements on a state-of-the-art model CoSeRec and propose Ti me I nterval Aware CoSeRec (TiCoSeRec). Experimental results on four datasets demonstrate that TiCoSeRec achieves significantly better performance than other 11 counterparts recommendation techniques.
27 October 2020 in IEEE Transactions on Neural Networks and Learning Systems
Negative sampling plays an important role in ranking-based recommender models. However, most existing sampling methods cannot generate informative item pairs with positive and negative instances due to two limitations: 1) they merely treat observed items as positive instances, ignoring the existence of potential positive items (i.e., nonobserved items users may prefer) and the probability of observed but noisy items and 2) they fail to capture the relationship between positive and negative items during negative sampling, which may cause the unexpected selection of potential positive items. In this article, we introduce a dynamic sampling strategy to search informative item pairs. Specifically, we first sample a positive instance from all the items by leveraging the overall features of user's observed items. Then, we strategically select a negative instance by considering its correlation with the sampled positive one. Formally, we propose an item pair generative adversarial network named IPGAN, where our sampling strategy is realized in two generative models for positive and negative instances, respectively. In addition, IPGAN can also ensure that the sampled item pairs are informative relative to the ground truth by a discriminative model. What is more, we propose a batch-training approach to further enhance both user and item modeling by alleviating the special bias (noise) from different users. This approach can also significantly accelerate the process of model training compared with classical GAN method for recommendation. Experimental results on three real data sets show that our approach outperforms other state-of-the-art approaches in terms of recommendation accuracy.
The goal of continual learning (CL) is to continuously learn new tasks without forgetting previously learned old tasks. To alleviate catastrophic forgetting, gradient projection based CL methods require that the gradient updates of new tasks are orthogonal to the subspace spanned by old tasks. This limits the learning process and leads to poor performance on the new task due to the projection constraint being too strong. In this paper, we first revisit the gradient projection method from the perspective of flatness of loss surface, and find that unflatness of the loss surface leads to catastrophic forgetting of the old tasks when the projection constraint is reduced to improve the performance of new tasks. Based on our findings, we propose a Data Augmented Flatness-aware Gradient Projection (DFGP) method to solve the problem, which consists of three modules: data and weight perturbation, flatness-aware optimization, and gradient projection. Specifically, we first perform a flatness-aware perturbation on the task data and current weights to find the case that makes the task loss worst. Next, flatness-aware optimization optimizes both the loss and the flatness of the loss surface on raw and worst-case perturbed data to obtain a flatness-aware gradient. Finally, gradient projection updates the network with the flatness-aware gradient along directions orthogonal to the subspace of the old tasks. Extensive experiments on four datasets show that our method improves the flatness of loss surface and the performance of new tasks, and achieves state-of-the-art (SOTA) performance in the average accuracy of all tasks.
12 April 2024 in ACM Transactions on Knowledge Discovery From Data
Multi-scenario and multi-task recommendation can use various feedback behaviors of users in different scenarios to learn users’ preferences and then make recommendations, which has attracted attention. However, the existing work ignores feature interactions and the fact that a pair of feature interactions will have differing levels of importance under different scenario-task pairs, leading to sub-optimal user preference learning. In this article, we propose a M ulti-scenario and M ulti-task aware F eature I nteraction model, dubbed MMFI , to explicitly model feature interactions and learn the importance of feature interaction pairs in different scenarios and tasks. Specifically, MMFI first incorporates a pairwise feature interaction unit and a scenario-task interaction unit to effectively capture the interaction of feature pairs and scenario-task pairs. Then MMFI designs a scenario-task aware attention layer for learning the importance of feature interactions from coarse-grained to fine-grained, improving the model’s performance on various scenario-task pairs. More specifically, this attention layer consists of three modules: a fully shared bottom module, a partially shared middle module, and a specific output module. Finally, MMFI adapts two sparsity-aware functions to remove some useless feature interactions. Extensive experiments on two public datasets demonstrate the superiority of the proposed method over the existing multi-task recommendation, multi-scenario recommendation, and multi-scenario & multi-task recommendation models.