Assessing Pre-trained Models for Transfer Learning through Distribution of Spectral Components

Pre-trained model assessment for transfer learning aims to identify the optimal candidate for the downstream tasks from a model hub, without the need of time-consuming fine-tuning. Existing advanced works mainly focus on analyzing the intrinsic characteristics of the entire features extracted by each pre-trained model or how well such features fit the target labels. This paper proposes a novel perspective for pre-trained model assessment through the Distribution of Spectral Components (DISCO). Through singular value decomposition of features extracted from pre-trained models, we investigate different spectral components and observe that they possess distinct transferability, contributing diversely to the fine-tuning performance. Inspired by this, we propose an assessment method based on the distribution of spectral components which measures the proportions of their corresponding singular values. Pre-trained models with features concentrating on more transferable components are regarded as better choices for transfer learning. We further leverage the labels of downstream data to better estimate the transferability of each spectral component and derive the final assessment criterion. Our proposed method is flexible and can be applied to both classification and regression tasks. We conducted comprehensive experiments across three benchmarks and two tasks including image classification and object detection, demonstrating that our method achieves state-of-the-art performance in choosing proper pre-trained models from the model hub for transfer learning.

EasyTime: Time Series Forecasting Made Easy

Time series forecasting has important applications across diverse domains. EasyTime, the system we demonstrate, facilitates easy use of time-series forecasting methods by researchers and practitioners alike. First, EasyTime enables one-click evaluation, enabling researchers to evaluate new forecasting methods using the suite of diverse time series datasets collected in the preexisting time series forecasting benchmark (TFB). This is achieved by leveraging TFB's flexible and consistent evaluation pipeline. Second, when practitioners must perform forecasting on a new dataset, a nontrivial first step is often to find an appropriate forecasting method. EasyTime provides an Automated Ensemble module that combines the promising forecasting methods to yield superior forecasting accuracy compared to individual methods. Third, EasyTime offers a natural language Q&A module leveraging large language models. Given a question like "Which method is best for long term forecasting on time series with strong seasonality?", EasyTime converts the question into SQL queries on the database of results obtained by TFB and then returns an answer in natural language and charts. By demonstrating EasyTime, we intend to show how it is possible to simplify the use of time series forecasting and to offer better support for the development of new generations of time series forecasting methods.

MultiRC: Joint Learning for Time Series Anomaly Prediction and Detection with Multi-scale Reconstructive Contrast

Many methods have been proposed for unsupervised time series anomaly detection. Despite some progress, research on predicting future anomalies is still relatively scarce. Predicting anomalies is particularly challenging due to the diverse reaction time and the lack of labeled data. To address these challenges, we propose MultiRC to integrate reconstructive and contrastive learning for joint learning of anomaly prediction and detection, with multi-scale structure and adaptive dominant period mask to deal with the diverse reaction time. MultiRC also generates negative samples to provide essential training momentum for the anomaly prediction tasks and prevent model degradation. We evaluate seven benchmark datasets from different fields. For both anomaly prediction and detection tasks, MultiRC outperforms existing state-of-the-art methods.

FoundTS: Comprehensive and Unified Benchmarking of Foundation Models for Time Series Forecasting

Time Series Forecasting (TSF) is key functionality in numerous fields, including in finance, weather services, and energy management. While TSF methods are emerging these days, many of them require domain-specific data collection and model training and struggle with poor generalization performance on new domains. Foundation models aim to overcome this limitation. Pre-trained on large-scale language or time series data, they exhibit promising inferencing capabilities in new or unseen data. This has spurred a surge in new TSF foundation models. We propose a new benchmark, FoundTS, to enable thorough and fair evaluation and comparison of such models. FoundTS covers a variety of TSF foundation models, including those based on large language models and those pretrained on time series. Next, FoundTS supports different forecasting strategies, including zero-shot, few-shot, and full-shot, thereby facilitating more thorough evaluations. Finally, FoundTS offers a pipeline that standardizes evaluation processes such as dataset splitting, loading, normalization, and few-shot sampling, thereby facilitating fair evaluations. Building on this, we report on an extensive evaluation of TSF foundation models on a broad range of datasets from diverse domains and with different statistical characteristics. Specifically, we identify pros and cons and inherent limitations of existing foundation models, and we identify directions for future model design. We make our code and datasets available at https://anonymous.4open.science/r/FoundTS-C2B0.

ROSE: Register Assisted General Time Series Forecasting with Decomposed Frequency Learning

With the increasing collection of time series data from various domains, there arises a strong demand for general time series forecasting models pre-trained on a large number of time-series datasets to support a variety of downstream prediction tasks. Enabling general time series forecasting faces two challenges: how to obtain unified representations from multi-domian time series data, and how to capture domain-specific features from time series data across various domains for adaptive transfer in downstream tasks. To address these challenges, we propose a Register Assisted General Time Series Forecasting Model with Decomposed Frequency Learning (ROSE), a novel pre-trained model for time series forecasting. ROSE employs Decomposed Frequency Learning for the pre-training task, which decomposes coupled semantic and periodic information in time series with frequency-based masking and reconstruction to obtain unified representations across domains. We also equip ROSE with a Time Series Register, which learns to generate a register codebook to capture domain-specific representations during pre-training and enhances domain-adaptive transfer by selecting related register tokens on downstream tasks. After pre-training on large-scale time series data, ROSE achieves state-of-the-art forecasting performance on 8 real-world benchmarks. Remarkably, even in few-shot scenarios, it demonstrates competitive or superior performance compared to existing methods trained with full data.

Pathformer: Multi-scale transformers with Adaptive Pathways for Time Series Forecasting

Transformer-based models have achieved some success in time series forecasting. Existing methods mainly model time series from limited or fixed scales, making it challenging to capture different characteristics spanning various scales. In this paper, we propose multi-scale transformers with adaptive pathways (Pathformer). The proposed Transformer integrates both temporal resolution and temporal distance for multi-scale modeling. Multi-scale division divides the time series into different temporal resolutions using patches of various sizes. Based on the division of each scale, dual attention is performed over these patches to capture global correlations and local details as temporal dependencies. We further enrich the multi-scale transformer with adaptive pathways, which adaptively adjust the multi-scale modeling process based on the varying temporal dynamics in the input time series, improving the prediction accuracy and generalization of Pathformer. Extensive experiments on eleven real-world datasets demonstrate that Pathformer not only achieves state-of-the-art performance by surpassing all current models but also exhibits stronger generalization abilities under various transfer scenarios.

CLIPood: Generalizing CLIP to Out-of-Distributions

Out-of-distribution (OOD) generalization, where the model needs to handle distribution shifts from training, is a major challenge of machine learning. Recently, contrastive language-image pre-training (CLIP) models have shown impressive zero-shot ability, revealing a promising path toward OOD generalization. However, to boost upon zero-shot performance, further adaptation of CLIP on downstream tasks is indispensable but undesirably degrades OOD generalization ability. In this paper, we aim at generalizing CLIP to out-of-distribution test data on downstream tasks. Beyond the two canonical OOD situations, domain shift and open class, we tackle a more general but difficult in-the-wild setting where both OOD situations may occur on the unseen test data. We propose CLIPood, a simple fine-tuning method that can adapt CLIP models to all OOD situations. To exploit semantic relations between classes from the text modality, CLIPood introduces a new training objective, margin metric softmax (MMS), with class adaptive margins for fine-tuning. Moreover, to incorporate both the pre-trained zero-shot model and the fine-tuned task-adaptive model, CLIPood proposes a new Beta moving average (BMA) to maintain a temporal ensemble according to Beta distribution. Experiments on diverse datasets with different OOD scenarios show that CLIPood consistently outperforms existing generalization techniques.

Hub-Pathway: Transfer Learning from A Hub of Pre-trained Models

Transfer learning aims to leverage knowledge from pre-trained models to benefit the target task. Prior transfer learning work mainly transfers from a single model. However, with the emergence of deep models pre-trained from different resources, model hubs consisting of diverse models with various architectures, pre-trained datasets and learning paradigms are available. Directly applying single-model transfer learning methods to each model wastes the abundant knowledge of the model hub and suffers from high computational cost. In this paper, we propose a Hub-Pathway framework to enable knowledge transfer from a model hub. The framework generates data-dependent pathway weights, based on which we assign the pathway routes at the input level to decide which pre-trained models are activated and passed through, and then set the pathway aggregation at the output level to aggregate the knowledge from different models to make predictions. The proposed framework can be trained end-to-end with the target task-specific loss, where it learns to explore better pathway configurations and exploit the knowledge in pre-trained models for each target datum. We utilize a noisy pathway generator and design an exploration loss to further explore different pathways throughout the model hub. To fully exploit the knowledge in pre-trained models, each model is further trained by specific data that activate it, which ensures its performance and enhances knowledge transfer. Experiment results on computer vision and reinforcement learning tasks demonstrate that the proposed Hub-Pathway framework achieves the state-of-the-art performance for model hub transfer learning.

Transferability in Deep Learning: A Survey

The success of deep learning algorithms generally depends on large-scale data, while humans appear to have inherent ability of knowledge transfer, by recognizing and applying relevant knowledge from previous learning experiences when encountering and solving unseen tasks. Such an ability to acquire and reuse knowledge is known as transferability in deep learning. It has formed the long-term quest towards making deep learning as data-efficient as human learning, and has been motivating fruitful design of more powerful deep learning algorithms. We present this survey to connect different isolated areas in deep learning with their relation to transferability, and to provide a unified and complete view to investigating transferability through the whole lifecycle of deep learning. The survey elaborates the fundamental goals and challenges in parallel with the core principles and methods, covering recent cornerstones in deep architectures, pre-training, task adaptation and domain adaptation. This highlights unanswered questions on the appropriate objectives for learning transferable knowledge and for adapting the knowledge to new tasks and domains, avoiding catastrophic forgetting and negative transfer. Finally, we implement a benchmark and an open-source library, enabling a fair evaluation of deep learning methods in terms of transferability.

Omni-Training for Data-Efficient Deep Learning

Learning a generalizable deep model from a few examples in a short time remains a major challenge of machine learning, which has impeded its wide deployment to many scenarios. Recent advances reveal that a properly pre-trained model endows an important property: transferability. A higher transferability of the learned representations indicates a better generalizability across domains of different distributions (domain transferability), or across tasks of different semantics (task transferability). Transferability has become the key to enable data-efficient deep learning, however, existing pre-training methods focus only on the domain transferability while meta-training methods only on the task transferability. This restricts their data-efficiency in downstream scenarios of diverging domains and tasks. A finding of this paper is that even a tight combination of pre-training and meta-training cannot achieve both kinds of transferability. This motivates the proposed Omni-Training framework towards data-efficient deep learning. Our first contribution is Omni-Net, a tri-flow architecture. Besides the joint representation flow, Omni-Net introduces two new parallel flows for pre-training and meta-training, respectively responsible for learning representations of domain transferability and task transferability. Omni-Net coordinates the parallel flows by routing them via the joint-flow, making each gain the other kind of transferability. Our second contribution is Omni-Loss, in which a mean-teacher regularization is imposed to learn generalizable and stabilized representations. Omni-Training is a general framework that accommodates many existing pre-training and meta-training algorithms. A thorough evaluation on cross-task and cross-domain datasets in classification, regression and reinforcement learning problems shows that Omni-Training consistently outperforms the state-of-the-art methods.