Continuous Domain Adaptation with Variational Domain-Agnostic Feature Replay
Imagia；Montreal Institute for Learning Algorithms (MILA), University of Montreal（蒙特利尔学习算法研究所（MILA），蒙特利尔大学）；Dalhousie University（达尔豪西大学）；Canadian Institute for Advanced Research (CIFAR)（加拿大高级研究所（CIFAR））
Qicheng Lao, Xiang Jiang, Mohammad Havaei, Yoshua Bengio
Learning in non-stationary environments is one of the biggest challenges in machine learning. Non-stationarity can be caused by either task drift, i.e., the drift in the conditional distribution of labels given the input data, or the domain drift, i.e., the drift in the marginal distribution of the input data. This paper aims to tackle this challenge in the context of continuous domain adaptation, where the model is required to learn new tasks adapted to new domains in a non-stationary environment while maintaining previously learned knowledge. To deal with both drifts, we propose variational domain-agnostic feature replay, an approach that is composed of three components: an inference module that filters the input data into domain-agnostic representations, a generative module that facilitates knowledge transfer, and a solver module that applies the filtered and transferable knowledge to solve the queries. We address the two fundamental scenarios in continuous domain adaptation, demonstrating the effectiveness of our proposed approach for practical usage.
Indian Institute of Technology Bombay（印度理工学院孟买分校）；Goethe University Frankfurt（法兰克福歌德大学）
We tackle the novel problem of incremental unsupervised domain adaptation (IDA) in this paper. We assume that a labeled source domain and different unlabeled target domains are incrementally observed with the constraint that data corresponding to the current domain is only available at a time. The goal is to preserve the accuracies for all the past domains while generalizing well for the current domain. The IDA setup suffers due to the abrupt differences among the domains and the unavailability of past data including the source domain. Inspired by the notion of generative feature replay, we pr…
Technische Universität Darmstadt（达姆施塔特工业大学）；Max-Planck Institute for Intelligent Systems（马克斯·普朗克智能系统研究所）；Universität zu Lübeck（吕贝克大学）
Autonomous robots need to interact with unknown, unstructured and changing environments, constantly facing novel challenges. Therefore, continuous online adaptation for lifelong-learning and the need of sample-efficient mechanisms to adapt to changes in the environment, the constraints, the tasks, or the robot itself are crucial. In this work, we propose a novel framework for probabilistic online motion planning with online adaptation based on a bio-inspired stochastic recurrent neural network. By using learning signals which mimic the intrinsic motivation signalcognitive dissonance in additio…
IBM Research Europe（IBM欧洲研究院）；ETH Zurich（苏黎世联邦理工学院）；Uppsala University（乌普萨拉大学）
Deep learning models can achieve high accuracy when trained on large amounts of labeled data. However, real-world scenarios often involve several challenges: Training data may become available in installments, may originate from multiple different domains, and may not contain labels for training. Certain settings, for instance medical applications, often involve further restrictions that prohibit retention of previously seen data due to privacy regulations. In this work, to address such challenges, we study unsupervised segmentation in continual learning scenarios that involve domain shift. To…
STAMP: Outlier-Aware Test-Time Adaptation with Stable Memory Replay
Institute of Automation, Chinese Academy of Sciences（中国科学院自动化研究所）；School of Artificial Intelligence, University of Chinese Academy of Sciences（中国科学院大学人工智能学院）；University of Science and Technology of China（中国科学技术大学）
The knowledge replay technique has been widely used in many tasks such as continual learning and continuous domain adaptation. The key lies in how to effectively encode the knowledge extracted from previous data and replay them during current training procedure. A simple yet effective model to achieve knowledge replay is autoencoder. However, the number of stored latent codes in autoencoder increases linearly with the scale of data and the trained encoder is redundant for the replaying stage. In this paper, we propose a novel and efficient knowledge recording network (KRNet) which directly map…
The capacity of meta-learning algorithms to quickly adapt to a variety of tasks, including ones they did not experience during meta-training, has been a key factor in the recent success of these methods on few-shot learning problems. This particular advantage of using meta-learning over standard supervised or reinforcement learning is only well founded under the assumption that the adaptation phase does improve the performance of our model on the task of interest. However, in the classical framework of meta-learning, this constraint is only mildly enforced, if not at all, and we only see an im…
Many meta-learning algorithms can be formulated into an interleaved process, in the sense that task-specific predictors are learned during inner-task adaptation and meta-parameters are updated during meta-update. The normal meta-training strategy needs to differentiate through the inner-task adaptation procedure to optimize the meta-parameters. This leads to a constraint that the inner-task algorithms should be solved analytically. Under this constraint, only simple algorithms with analytical solutions can be applied as the inner-task algorithms, limiting the model expressiveness. To lift the …
We propose a normalization layer for unsupervised domain adaption in semantic scene segmentation. Normalization layers are known to improve convergence and generalization and are part of many state-of-the-art fully-convolutional neural networks. We show that conventional normalization layers worsen the performance of current Unsupervised Adversarial Domain Adaption (UADA), which is a method to improve network performance on unlabeled datasets and the focus of our research. Therefore, we propose a novel Domain Agnostic Normalization layer and thereby unlock the benefits of normalization layers …
NLPR, Institute of Automation, Chinese Academy of Sciences（中国科学院自动化研究所模式识别国家重点实验室）；School of Artificial Intelligence, University of Chinese Academy of Sciences（中国科学院大学人工智能学院）；Youtu Lab, Tencent Inc.（腾讯优图实验室）；CASIA-LLvision Joint Lab（中国科学院自动化研究所与联影智能联合实验室）；The Chinese University of Hong Kong, Shenzhen（香港中文大学（深圳））；Shenzhen Research Institute of Big Data（深圳市大数据研究院）
Big progress has been achieved in domain adaptation in decades. Existing works are always based on an ideal assumption that testing target domain are i.i.d. with training target domains. However, due to unpredictable corruptions (e.g., noise and blur) in real data like web images, domain adaptation methods are increasingly required to be corruption robust on target domains. In this paper, we investigate a new task, Corruption-agnostic Robust Domain Adaptation (CRDA): to be accurate on original data and robust against unavailable-for-training corruptions on target domains. This task is non-triv…
JD AI Research（京东人工智能研究院）；City University of Hong Kong（香港城市大学）
Unsupervised domain adaptation has received significant attention in recent years. Most of existing works tackle the closed-set scenario, assuming that the source and target domains share the exactly same categories. In practice, nevertheless, a target domain often contains samples of classes unseen in source domain (i.e., unknown class). The extension of domain adaptation from closed-set to such open-set situation is not trivial since the target samples in unknown class are not expected to align with the source. In this paper, we address this problem by augmenting the state-of-the-art domain …
Janelia Research Campus, HHMI（霍华德·休斯医学研究所珍妮莉亚研究园区）；Harvard University（哈佛大学）；Columbia University（哥伦比亚大学）
Meta-learning has emerged as an important framework for learning new tasks from just a few examples. The success of any meta-learning model depends on (i) its fast adaptation to new tasks, as well as (ii) having a shared representation across similar tasks. Here we extend the model-agnostic meta-learning (MAML) framework introduced by Finn et al. (2017) to achieve improved performance by analyzing the temporal dynamics of the optimization procedure via the Runge-Kutta method. This method enables us to gain fine-grained control over the optimization and helps us achieve both the adaptation and …
Computer Vision Center（计算机视觉中心）；Universitat Autònoma de Barcelona（巴塞罗那自治大学）
Online continual learning aims to learn from a non-IID stream of data from a number of different tasks, where the learner is only allowed to consider data once. Methods are typically allowed to use a limited buffer to store some of the images in the stream. Recently, it was found that feature replay, where an intermediate layer representation of the image is stored (or generated) leads to superior results than image replay, while requiring less memory. Quantized exemplars can further reduce the memory usage. However, a drawback of these methods is that they use a fixed (or very intransigent) b…
Artificial Intelligence, Analytics And Informatics (AI), A*STAR（A*STAR人工智能、分析与信息学）；RMIT University（皇家墨尔本理工大学）；Institute for Infocomm Research (IR), A*STAR（A*STAR资讯通信研究院）
Deep neural networks have shown promise in several domains, and the learned data (task) specific information is implicitly stored in the network parameters. Extraction and utilization of encoded knowledge representations are vital when data is no longer available in the future, especially in a continual learning scenario. In this work, we introduce {\em flashcards}, which are visual representations that {\em capture} the encoded knowledge of a network as a recursive function of predefined random image patterns. In a continual learning scenario, flashcards help to prevent catastrophic forgettin…
The Australian Institute for Machine Learning（澳大利亚机器学习研究所）；The University of Adelaide（阿德莱德大学）；The University of New South Wales（新南威尔士大学）
Continual Learning (CL) methods aim to enable machine learning models to learn new tasks without catastrophic forgetting of those that have been previously mastered. Existing CL approaches often keep a buffer of previously-seen samples, perform knowledge distillation, or use regularization techniques towards this goal. Despite their performance, they still suffer from interference across tasks which leads to catastrophic forgetting. To ameliorate this problem, we propose to only activate and select sparse neurons for learning current and past tasks at any stage. More parameters space and model…
We propose a method for unsupervised domain adaptation that trains a shared embedding to align the joint distributions of inputs (domain) and outputs (classes), making any classifier agnostic to the domain. Joint alignment ensures that not only the marginal distributions of the domain are aligned, but the labels as well. We propose a novel objective function that encourages the class-conditional distributions to have disjoint support in feature space. We further exploit adversarial regularization to improve the performance of the classifier on the domain for which no annotated data is availabl…
University of California, Berkeley（加州大学伯克利分校）；OpenAI
We propose an algorithm for meta-learning that is model-agnostic, in the sense that it is compatible with any model trained with gradient descent and applicable to a variety of different learning problems, including classification, regression, and reinforcement learning. The goal of meta-learning is to train a model on a variety of learning tasks, such that it can solve new learning tasks using only a small number of training samples. In our approach, the parameters of the model are explicitly trained such that a small number of gradient steps with a small amount of training data from a new ta…
Huawei Noah’s Ark Lab（华为诺亚方舟实验室）；University of Rome Sapienza（罗马第一大学）；Politecnico di Torino（都灵理工大学）；Italian Institute of Technology（意大利技术研究院）
The ability to generalize across visual domains is crucial for the robustness of artificial recognition systems. Although many training sources may be available in real contexts, the access to even unlabeled target samples cannot be taken for granted, which makes standard unsupervised domain adaptation methods inapplicable in the wild. In this work we investigate how to exploit multiple sources by hallucinating a deep visual domain composed of images, possibly unrealistic, able to maintain categorical knowledge while discarding specific source styles. The produced agnostic images are the resul…
ChainLoRA: Geometry-Preserving Task Vector Merging for Continual Learning in LLMs
Shanghai Jiao Tong University（上海交通大学）；Shanghai Innovation Institute（上海创新研究院）；Beijing Zhongguancun Academy（北京中关村学院）；StepFun（阶跃星辰）；Huazhong University of Science and Technology（华中科技大学）；IAAR（智能自动化研究院）
Speaker adaptive training (SAT) of neural network acoustic models learns models in a way that makes them more suitable for adaptation to test conditions. Conventionally, model-based speaker adaptive training is performed by having a set of speaker dependent parameters that are jointly optimised with speaker independent parameters in order to remove speaker variation. However, this does not scale well if all neural network weights are to be adapted to the speaker. In this paper we formulate speaker adaptive training as a meta-learning task, in which an adaptation process using gradient descent …
TADA: Efficient Task-Agnostic Domain Adaptation for Transformers
NEC Laboratories Europe GmbH（NEC欧洲实验室有限公司）；University of Mannheim（曼海姆大学）；Bosch Center for Artificial Intelligence（博世人工智能中心）；Karlsruhe University of Applied Sciences（卡尔斯鲁厄应用技术大学）
Cornell University（康奈尔大学）；the Ohio State University（俄亥俄州立大学）
Self-driving cars must detect other vehicles and pedestrians in 3D to plan safe routes and avoid collisions. State-of-the-art 3D object detectors, based on deep learning, have shown promising accuracy but are prone to over-fit to domain idiosyncrasies, making them fail in new environments -- a serious problem if autonomous vehicles are meant to operate freely. In this paper, we propose a novel learning approach that drastically reduces this gap by fine-tuning the detector on pseudo-labels in the target domain, which our method generates while the vehicle is parked, based on replays of previous…
Modern machine learning systems need to be able to cope with constantly arriving and changing data. Two main areas of research dealing with such scenarios are continual learning and data stream mining. Continual learning focuses on accumulating knowledge and avoiding forgetting, assuming information once learned should be stored. Data stream mining focuses on adaptation to concept drift and discarding outdated information, assuming that only the most recent data is relevant. While these two areas are mainly being developed in separation, they offer complementary views on the problem of learnin…
New York University（纽约大学）；DeepMind（深度思维）；Google Research（谷歌研究院）
Learning new tasks continuously without forgetting on a constantly changing data distribution is essential for real-world problems but extremely challenging for modern deep learning. In this work we propose HCL, a Hybrid generative-discriminative approach to Continual Learning for classification. We model the distribution of each task and each class with a normalizing flow. The flow is used to learn the data distribution, perform classification, identify task changes, and avoid forgetting, all leveraging the invertibility and exact likelihood which are uniquely enabled by the normalizing flow …
Rensselaer Polytechnic Institute（伦斯勒理工学院）；University of Michigan（密歇根大学）；MIT-IBM Watson AI Lab（麻省理工学院-IBM沃森人工智能实验室）；IBM Research（IBM研究院）；Northeastern University（东北大学）；Michigan State University（密歇根州立大学）
Model-agnostic meta-learning (MAML) has emerged as one of the most successful meta-learning techniques in few-shot learning. It enables us to learn a meta-initialization} of model parameters (that we call meta-model) to rapidly adapt to new tasks using a small amount of labeled training data. Despite the generalization power of the meta-model, it remains elusive that how adversarial robustness can be maintained by MAML in few-shot learning. In addition to generalization, robustness is also desired for a meta-model to defend adversarial examples (attacks). Toward promoting adversarial robustnes…
Continual Learning with Elastic Regularization and Synthetic Replay for Federated MLLM Fine-Tuning
The University of British Columbia（英属哥伦比亚大学）；Fudan University（复旦大学）；Royal College of Science, Imperial College London（伦敦帝国理工学院皇家科学学院）；Dyson School of Design Engineering（戴森设计工程学院）；Suzhou Institute of Biomedical Engineering and Technology (SIBET), Chinese Academy of Sciences（中国科学院苏州生物医学工程技术研究所）；East China Normal University（华东师范大学）
University of Central Florida（中佛罗里达大学）；University of Macau（澳门大学）；Nanjing University（南京大学）
In many scenarios, 1) data streams are generated in real time; 2) labeled data are expensive and only limited labels are available in the beginning; 3) real-world data is not always i.i.d. and data drift over time gradually; 4) the storage of historical streams is limited and model updating can only be achieved based on a very short lookback window. This learning setting limits the applicability and availability of many Machine Learning (ML) algorithms. We generalize the learning task under such setting as a semi-supervised drifted stream learning with short lookback problem (SDSL). SDSL impos…
Training a multi-speaker Text-to-Speech (TTS) model from scratch is computationally expensive and adding new speakers to the dataset requires the model to be re-trained. The naive solution of sequential fine-tuning of a model for new speakers can lead to poor performance of older speakers. This phenomenon is known as catastrophic forgetting. In this paper, we look at TTS modeling from a continual learning perspective, where the goal is to add new speakers without forgetting previous speakers. Therefore, we first propose an experimental setup and show that serial fine-tuning for new speakers ca…
In recent years, machine learning has achieved impressive results across different application areas. However, machine learning algorithms do not necessarily perform well on a new domain with a different distribution than its training set. Domain Adaptation (DA) is used to mitigate this problem. One approach of existing DA algorithms is to find domain invariant features whose distributions in the source domain are the same as their distribution in the target domain. In this paper, we propose to let the classifier that performs the final classification task on the target domain learn implicitly…
Swiss Federal Institute of Technology Lausanne (EPFL)（洛桑联邦理工学院）
Session-based recommendation has received growing attention recently due to the increasing privacy concern. Despite the recent success of neural session-based recommenders, they are typically developed in an offline manner using a static dataset. However, recommendation requires continual adaptation to take into account new and obsolete items and users, and requires "continual learning" in real-life applications. In this case, the recommender is updated continually and periodically with new data that arrives in each update cycle, and the updated model needs to provide recommendations for user …
