Since the recent success of Vision Transformers (ViTs), explorations toward transformer-style architectures have triggered the resurgence of modern ConvNets. In this work, we explore the representation ability of DNNs through the lens of interaction complexities. We empirically show that interaction complexity is an overlooked but essential indicator for visual recognition. Accordingly, a new family of efficient ConvNets, named MogaNet, is presented to pursue informative context mining in pure ConvNet-based models, with preferable complexity-performance trade-offs. In MogaNet, interactions across multiple complexities are facilitated and contextualized by leveraging two specially designed aggregation blocks in both spatial and channel interaction spaces. Extensive studies are conducted on ImageNet classification, COCO object detection, and ADE20K semantic segmentation tasks. The results demonstrate that our MogaNet establishes new state-of-the-art over other popular methods in mainstream scenarios and all model scales. Typically, the lightweight MogaNet-T achieves 80.0\% top-1 accuracy with only 1.44G FLOPs using a refined training setup on ImageNet-1K, surpassing ParC-Net-S by 1.4\% accuracy but saving 59\% (2.04G) FLOPs.
Recently, the unified streaming and non-streaming two-pass (U2/U2++) end-to-end model for speech recognition has shown great performance in terms of streaming capability, accuracy and latency. In this paper, we present fast-U2++, an enhanced version of U2++ to further reduce partial latency. The core idea of fast-U2++ is to output partial results of the bottom layers in its encoder with a small chunk, while using a large chunk in the top layers of its encoder to compensate the performance degradation caused by the small chunk. Moreover, we use knowledge distillation method to reduce the token emission latency. We present extensive experiments on Aishell-1 dataset. Experiments and ablation studies show that compared to U2++, fast-U2++ reduces model latency from 320ms to 80ms, and achieves a character error rate (CER) of 5.06% with a streaming setup.
In this paper, we present TrimTail, a simple but effective emission regularization method to improve the latency of streaming ASR models. The core idea of TrimTail is to apply length penalty (i.e., by trimming trailing frames, see Fig. 1-(b)) directly on the spectrogram of input utterances, which does not require any alignment. We demonstrate that TrimTail is computationally cheap and can be applied online and optimized with any training loss or any model architecture on any dataset without any extra effort by applying it on various end-to-end streaming ASR networks either trained with CTC loss [1] or Transducer loss [2]. We achieve 100 $\sim$ 200ms latency reduction with equal or even better accuracy on both Aishell-1 and Librispeech. Moreover, by using TrimTail, we can achieve a 400ms algorithmic improvement of User Sensitive Delay (USD) with an accuracy loss of less than 0.2.
The recently proposed Conformer architecture which combines convolution with attention to capture both local and global dependencies has become the \textit{de facto} backbone model for Automatic Speech Recognition~(ASR). Inherited from the Natural Language Processing (NLP) tasks, the architecture takes Layer Normalization~(LN) as a default normalization technique. However, through a series of systematic studies, we find that LN might take 10\% of the inference time despite that it only contributes to 0.1\% of the FLOPs. This motivates us to replace LN with other normalization techniques, e.g., Batch Normalization~(BN), to speed up inference with the help of operator fusion methods and the avoidance of calculating the mean and variance statistics during inference. After examining several plain attempts which directly remove all LN layers or replace them with BN in the same place, we find that the divergence issue is mainly caused by the unstable layer output. We therefore propose to append a BN layer to each linear or convolution layer where stabilized training results are observed. We also propose to simplify the activations in Conformer, such as Swish and GLU, by replacing them with ReLU. All these exchanged modules can be fused into the weights of the adjacent linear/convolution layers and hence have zero inference cost. Therefore, we name it FusionFormer. Our experiments indicate that FusionFormer is as effective as the LN-based Conformer and is about 10\% faster.
The building sector has been recognized as one of the primary sectors for worldwide energy consumption. Improving the energy efficiency of the building sector can help reduce the operation cost and reduce the greenhouse gas emission. The energy management system (EMS) can monitor and control the operations of built-in appliances in buildings, so an efficient EMS is of crucial importance to improve the building operation efficiency and maintain safe operations. With the growing penetration of renewable energy and electrical appliances, increasing attention has been paid to the development of intelligent building EMS. Recently, reinforcement learning (RL) has been applied for building EMS and has shown promising potential. However, most of the current RL-based EMS solutions would need a large amount of data to learn a reliable control policy, which limits the applicability of these solutions in the real world. In this work, we propose MetaEMS, which can help achieve better energy management performance with the benefits of RL and meta-learning. Experiment results showcase that our proposed MetaEMS can adapt faster to environment changes and perform better in most situations compared with other baselines.
Machine learning on electromyography (EMG) has recently achieved remarkable success on a variety of tasks, while such success relies heavily on the assumption that the training and future data must be of the same data distribution. However, this assumption may not hold in many real-world applications. Model calibration is required via data re-collection and label annotation, which is generally very expensive and time-consuming. To address this problem, transfer learning (TL), which aims to improve target learners' performance by transferring the knowledge from related source domains, is emerging as a new paradigm to reduce the amount of calibration effort. In this survey, we assess the eligibility of more than fifty published peer-reviewed representative transfer learning approaches for EMG applications. Unlike previous surveys on purely transfer learning or EMG-based machine learning, this survey aims to provide an insight into the biological foundations of existing transfer learning methods on EMG-related analysis. In specific, we first introduce the physiological structure of the muscles and the EMG generating mechanism, and the recording of EMG to provide biological insights behind existing transfer learning approaches. Further, we categorize existing research endeavors into data based, model based, training scheme based, and adversarial based. This survey systematically summarizes and categorizes existing transfer learning approaches for EMG related machine learning applications. In addition, we discuss possible drawbacks of existing works and point out the future direction of better EMG transfer learning algorithms to enhance practicality for real-world applications.
With the remarkable progress of deep neural networks in computer vision, data mixing augmentation techniques are widely studied to alleviate problems of degraded generalization when the amount of training data is limited. However, mixup strategies have not been well assembled in current vision toolboxes. In this paper, we propose \texttt{OpenMixup}, an open-source all-in-one toolbox for supervised, semi-, and self-supervised visual representation learning with mixup. It offers an integrated model design and training platform, comprising a rich set of prevailing network architectures and modules, a collection of data mixing augmentation methods as well as practical model analysis tools. In addition, we also provide standard mixup image classification benchmarks on various datasets, which expedites practitioners to make fair comparisons among state-of-the-art methods under the same settings. The source code and user documents are available at \url{https://github.com/Westlake-AI/openmixup}.
Federated learning (FL) has gained significant attention recently as a privacy-enhancing tool to jointly train a machine learning model by multiple participants. The prior work on FL has mostly studied how to protect label privacy during model training. However, model evaluation in FL might also lead to potential leakage of private label information. In this work, we propose an evaluation algorithm that can accurately compute the widely used AUC (area under the curve) metric when using the label differential privacy (DP) in FL. Through extensive experiments, we show our algorithms can compute accurate AUCs compared to the ground truth.
Federated learning (FL) facilitates multiple clients to jointly train a machine learning model without sharing their private data. However, Non-IID data of clients presents a tough challenge for FL. Existing personalized FL approaches rely heavily on the default treatment of one complete model as a basic unit and ignore the significance of different layers on Non-IID data of clients. In this work, we propose a new framework, federated model components self-attention (FedMCSA), to handle Non-IID data in FL, which employs model components self-attention mechanism to granularly promote cooperation between different clients. This mechanism facilitates collaboration between similar model components while reducing interference between model components with large differences. We conduct extensive experiments to demonstrate that FedMCSA outperforms the previous methods on four benchmark datasets. Furthermore, we empirically show the effectiveness of the model components self-attention mechanism, which is complementary to existing personalized FL and can significantly improve the performance of FL.
Missing data is an inevitable and common problem in data-driven intelligent transportation systems (ITS). In the past decade, scholars have done many research on the recovery of missing traffic data, however how to make full use of spatio-temporal traffic patterns to improve the recovery performance is still an open problem. Aiming at the spatio-temporal characteristics of traffic speed data, this paper regards the recovery of missing data as a matrix completion problem, and proposes a spatio-temporal traffic data completion method based on hidden feature analysis, which discovers spatio-temporal patterns and underlying structures from incomplete data to complete the recovery task. Therefore, we introduce spatial and temporal correlation to capture the main underlying features of each dimension. Finally, these latent features are applied to recovery traffic data through latent feature analysis. The experimental and evaluation results show that the evaluation criterion value of the model is small, which indicates that the model has better performance. The results show that the model can accurately estimate the continuous missing data.