SparseGrow: Addressing Growth-Induced Forgetting in Task-Agnostic Continual Learning

In continual learning (CL), model growth enhances adaptability over new data, improving knowledge retention for more tasks. However, improper model growth can lead to severe degradation of previously learned knowledge, an issue we name as growth-induced forgetting (GIFt), especially in task-agnostic CL using entire grown model for inference. Existing works, despite adopting model growth and random initialization for better adaptability, often fail to recognize the presence of GIFt caused by improper model growth. This oversight limits comprehensive control of forgetting and hinders full utilization of model growth. We are the first in CL to identify this issue and conduct an in-depth study on root cause of GIFt, where layer expansion stands out among model growth strategies, widening layers without affecting model functionality. Yet, direct adoption of layer expansion presents challenges. It lacks data-driven control and initialization of expanded parameters to balance adaptability and knowledge retention. This paper presents a novel SparseGrow approach to overcome the issue of GIFt while enhancing adaptability over new data. SparseGrow employs data-driven sparse layer expansion to control efficient parameter usage during growth, reducing GIFt from excessive growth and functionality changes. It also combines sparse growth with on-data initialization at training late-stage to create partially 0-valued expansions that fit learned distribution, enhancing retention and adaptability. To further minimize forgetting, freezing is applied by calculating the sparse mask, allowing data-driven preservation of important parameters. Through experiments across datasets with various settings, cases and task numbers, we demonstrate the necessity of layer expansion and showcase the effectiveness of SparseGrow in overcoming GIFt, highlighting its adaptability and knowledge retention for incremental tasks.

FedDistill: Global Model Distillation for Local Model De-Biasing in Non-IID Federated Learning

Federated Learning (FL) is a novel approach that allows for collaborative machine learning while preserving data privacy by leveraging models trained on decentralized devices. However, FL faces challenges due to non-uniformly distributed (non-iid) data across clients, which impacts model performance and its generalization capabilities. To tackle the non-iid issue, recent efforts have utilized the global model as a teaching mechanism for local models. However, our pilot study shows that their effectiveness is constrained by imbalanced data distribution, which induces biases in local models and leads to a 'local forgetting' phenomenon, where the ability of models to generalize degrades over time, particularly for underrepresented classes. This paper introduces FedDistill, a framework enhancing the knowledge transfer from the global model to local models, focusing on the issue of imbalanced class distribution. Specifically, FedDistill employs group distillation, segmenting classes based on their frequency in local datasets to facilitate a focused distillation process to classes with fewer samples. Additionally, FedDistill dissects the global model into a feature extractor and a classifier. This separation empowers local models with more generalized data representation capabilities and ensures more accurate classification across all classes. FedDistill mitigates the adverse effects of data imbalance, ensuring that local models do not forget underrepresented classes but instead become more adept at recognizing and classifying them accurately. Our comprehensive experiments demonstrate FedDistill's effectiveness, surpassing existing baselines in accuracy and convergence speed across several benchmark datasets.

A Multimodal Foundation Agent for Financial Trading: Tool-Augmented, Diversified, and Generalist

Financial trading is a crucial component of the markets, informed by a multimodal information landscape encompassing news, prices, and Kline charts, and encompasses diverse tasks such as quantitative trading and high-frequency trading with various assets. While advanced AI techniques like deep learning and reinforcement learning are extensively utilized in finance, their application in financial trading tasks often faces challenges due to inadequate handling of multimodal data and limited generalizability across various tasks. To address these challenges, we present FinAgent, a multimodal foundational agent with tool augmentation for financial trading. FinAgent's market intelligence module processes a diverse range of data-numerical, textual, and visual-to accurately analyze the financial market. Its unique dual-level reflection module not only enables rapid adaptation to market dynamics but also incorporates a diversified memory retrieval system, enhancing the agent's ability to learn from historical data and improve decision-making processes. The agent's emphasis on reasoning for actions fosters trust in its financial decisions. Moreover, FinAgent integrates established trading strategies and expert insights, ensuring that its trading approaches are both data-driven and rooted in sound financial principles. With comprehensive experiments on 6 financial datasets, including stocks and Crypto, FinAgent significantly outperforms 9 state-of-the-art baselines in terms of 6 financial metrics with over 36% average improvement on profit. Specifically, a 92.27% return (a 84.39% relative improvement) is achieved on one dataset. Notably, FinAgent is the first advanced multimodal foundation agent designed for financial trading tasks.

MGAS: Multi-Granularity Architecture Search for Effective and Efficient Neural Networks

Differentiable architecture search (DAS) revolutionizes neural architecture search (NAS) with time-efficient automation, transitioning from discrete candidate sampling and evaluation to differentiable super-net optimization and discretization. However, existing DAS methods either only conduct coarse-grained operation-level search or manually define the remaining ratios for fine-grained kernel-level and weight-level units, which fail to simultaneously optimize model size and model performance. Furthermore, these methods compromise search quality to reduce memory consumption. To tackle these issues, we introduce multi-granularity architecture search (MGAS), a unified framework which aims to comprehensively and memory-efficiently explore the multi-granularity search space to discover both effective and efficient neural networks. Specifically, we learn discretization functions specific to each granularity level to adaptively determine the remaining ratios according to the evolving architecture. This ensures an optimal balance among units of different granularity levels for different target model sizes. Considering the memory demands, we break down the super-net optimization and discretization into multiple sub-net stages. Nevertheless, the greedy nature of this approach may introduce bias in the early stages. To compensate for the bias, we propose progressive re-evaluation to allow for re-pruning and regrowing of previous units during subsequent stages. Extensive experiments on CIFAR-10, CIFAR-100 and ImageNet demonstrate that MGAS outperforms other state-of-the-art methods in achieving a better trade-off between model performance and model size.

Revisiting Parameter Reuse to Overcome Catastrophic Forgetting in Neural Networks

Neural networks tend to forget previously learned knowledge when continuously learning on datasets with varying distributions, a phenomenon known as catastrophic forgetting. More significant distribution shifts among datasets lead to more forgetting. Recently, parameter-isolation-based approaches have shown great potential in overcoming forgetting with significant distribution shifts. However, they suffer from poor generalization as they fix the neural path for each dataset during training and require dataset labels during inference. In addition, they do not support backward knowledge transfer as they prioritize past data over future ones. In this paper, we propose a new adaptive learning method, named AdaptCL, that fully reuses and grows on learned parameters to overcome catastrophic forgetting and allows the positive backward transfer without requiring dataset labels. Our proposed technique adaptively grows on the same neural path by allowing optimal reuse of frozen parameters. Besides, it uses parameter-level data-driven pruning to assign equal priority to the data. We conduct extensive experiments on MNIST Variants, DomainNet, and Food Freshness Detection datasets under different intensities of distribution shifts without requiring dataset labels. Results demonstrate that our proposed method is superior to alternative baselines in minimizing forgetting and enabling positive backward knowledge transfer.