Multi-class classification problems often have many semantically similar classes. For example, 90 of ImageNet's 1000 classes are for different breeds of dog. We should expect that these semantically similar classes will have similar parameter vectors, but the standard cross entropy loss does not enforce this constraint. We introduce the tree loss as a drop-in replacement for the cross entropy loss. The tree loss re-parameterizes the parameter matrix in order to guarantee that semantically similar classes will have similar parameter vectors. Using simple properties of stochastic gradient descent, we show that the tree loss's generalization error is asymptotically better than the cross entropy loss's. We then validate these theoretical results on synthetic data, image data (CIFAR100, ImageNet), and text data (Twitter).
This paper describes the BLCU-ICALL system used in the SemEval-2022 Task 1 Comparing Dictionaries and Word Embeddings, the Definition Modeling subtrack, achieving 1st on Italian, 2nd on Spanish and Russian, and 3rd on English and French. We propose a transformer-based multitasking framework to explore the task. The framework integrates multiple embedding architectures through the cross-attention mechanism, and captures the structure of glosses through a masking language model objective. Additionally, we also investigate a simple but effective model ensembling strategy to further improve the robustness. The evaluation results show the effectiveness of our solution. We release our code at: https://github.com/blcuicall/SemEval2022-Task1-DM.
In this paper, we delve into two key techniques in Semi-Supervised Object Detection (SSOD), namely pseudo labeling and consistency training. We observe that these two techniques currently neglect some important properties of object detection, hindering efficient learning on unlabeled data. Specifically, for pseudo labeling, existing works only focus on the classification score yet fail to guarantee the localization precision of pseudo boxes; For consistency training, the widely adopted random-resize training only considers the label-level consistency but misses the feature-level one, which also plays an important role in ensuring the scale invariance. To address the problems incurred by noisy pseudo boxes, we design Noisy Pseudo box Learning (NPL) that includes Prediction-guided Label Assignment (PLA) and Positive-proposal Consistency Voting (PCV). PLA relies on model predictions to assign labels and makes it robust to even coarse pseudo boxes; while PCV leverages the regression consistency of positive proposals to reflect the localization quality of pseudo boxes. Furthermore, in consistency training, we propose Multi-view Scale-invariant Learning (MSL) that includes mechanisms of both label- and feature-level consistency, where feature consistency is achieved by aligning shifted feature pyramids between two images with identical content but varied scales. On COCO benchmark, our method, termed PSEudo labeling and COnsistency training (PseCo), outperforms the SOTA (Soft Teacher) by 2.0, 1.8, 2.0 points under 1%, 5%, and 10% labelling ratios, respectively. It also significantly improves the learning efficiency for SSOD, e.g., PseCo halves the training time of the SOTA approach but achieves even better performance.
In this letter, we investigate time-domain channel estimation for wideband millimeter wave (mmWave) MIMO OFDM system. By transmitting frequency-domain pilot symbols as well as different beamforming vectors, we observe that the time-domain mmWave MIMO channels exhibit channel delay sparsity and especially block sparsity among different spatial directions. Then we propose a time-domain channel estimation exploiting block sparsity (TDCEBS) scheme, which always aims at finding the best nonzero block achieving the largest projection of the residue at each iterations. In particular, we evaluate the system performance using the QuaDRiGa which is recommended by 5G New Radio to generate wideband mmWave MIMO channels. The effectiveness of the proposed TDCEBS scheme is verified by the simulation results, as the proposed scheme outperforms the existing schemes.
Knowledge Distillation (KD) is a widely-used technology to inherit information from cumbersome teacher models to compact student models, consequently realizing model compression and acceleration. Compared with image classification, object detection is a more complex task, and designing specific KD methods for object detection is non-trivial. In this work, we elaborately study the behaviour difference between the teacher and student detection models, and obtain two intriguing observations: First, the teacher and student rank their detected candidate boxes quite differently, which results in their precision discrepancy. Second, there is a considerable gap between the feature response differences and prediction differences between teacher and student, indicating that equally imitating all the feature maps of the teacher is the sub-optimal choice for improving the student's accuracy. Based on the two observations, we propose Rank Mimicking (RM) and Prediction-guided Feature Imitation (PFI) for distilling one-stage detectors, respectively. RM takes the rank of candidate boxes from teachers as a new form of knowledge to distill, which consistently outperforms the traditional soft label distillation. PFI attempts to correlate feature differences with prediction differences, making feature imitation directly help to improve the student's accuracy. On MS COCO and PASCAL VOC benchmarks, extensive experiments are conducted on various detectors with different backbones to validate the effectiveness of our method. Specifically, RetinaNet with ResNet50 achieves 40.4% mAP in MS COCO, which is 3.5% higher than its baseline, and also outperforms previous KD methods.
The integration of Vector Quantised Variational AutoEncoder (VQ-VAE) with autoregressive models as generation part has yielded high-quality results on image generation. However, the autoregressive models will strictly follow the progressive scanning order during the sampling phase. This leads the existing VQ series models to hardly escape the trap of lacking global information. Denoising Diffusion Probabilistic Models (DDPM) in the continuous domain have shown a capability to capture the global context, while generating high-quality images. In the discrete state space, some works have demonstrated the potential to perform text generation and low resolution image generation. We show that with the help of a content-rich discrete visual codebook from VQ-VAE, the discrete diffusion model can also generate high fidelity images with global context, which compensates for the deficiency of the classical autoregressive model along pixel space. Meanwhile, the integration of the discrete VAE with the diffusion model resolves the drawback of conventional autoregressive models being oversized, and the diffusion model which demands excessive time in the sampling process when generating images. It is found that the quality of the generated images is heavily dependent on the discrete visual codebook. Extensive experiments demonstrate that the proposed Vector Quantised Discrete Diffusion Model (VQ-DDM) is able to achieve comparable performance to top-tier methods with low complexity. It also demonstrates outstanding advantages over other vectors quantised with autoregressive models in terms of image inpainting tasks without additional training.
The foundation model is not the last chapter of the model production pipeline. Transferring with few data in a general way to thousands of downstream tasks is becoming a trend of the foundation model's application. In this paper, we proposed a universal transfer framework: One to Transfer All (OTA) to transfer any Vision Foundation Model (VFM) to any downstream tasks with few downstream data. We first transfer a VFM to a task-specific model by Image Re-representation Fine-tuning (IRF) then distilling knowledge from a task-specific model to a deployed model with data produced by Downstream Image-Guided Generation (DIGG). OTA has no dependency on upstream data, VFM, and downstream tasks when transferring. It also provides a way for VFM researchers to release their upstream information for better transferring but not leaking data due to privacy requirements. Massive experiments validate the effectiveness and superiority of our methods in few data setting. Our code will be released.
Enormous waves of technological innovations over the past several years, marked by the advances in AI technologies, are profoundly reshaping the industry and the society. However, down the road, a key challenge awaits us, that is, our capability of meeting rapidly-growing scenario-specific demands is severely limited by the cost of acquiring a commensurate amount of training data. This difficult situation is in essence due to limitations of the mainstream learning paradigm: we need to train a new model for each new scenario, based on a large quantity of well-annotated data and commonly from scratch. In tackling this fundamental problem, we move beyond and develop a new learning paradigm named INTERN. By learning with supervisory signals from multiple sources in multiple stages, the model being trained will develop strong generalizability. We evaluate our model on 26 well-known datasets that cover four categories of tasks in computer vision. In most cases, our models, adapted with only 10% of the training data in the target domain, outperform the counterparts trained with the full set of data, often by a significant margin. This is an important step towards a promising prospect where such a model with general vision capability can dramatically reduce our reliance on data, thus expediting the adoption of AI technologies. Furthermore, revolving around our new paradigm, we also introduce a new data system, a new architecture, and a new benchmark, which, together, form a general vision ecosystem to support its future development in an open and inclusive manner.
This paper presents a hierarchical segment-based optimization method for Simultaneous Localization and Mapping (SLAM) system. First we propose a reliable trajectory segmentation method that can be used to increase efficiency in the back-end optimization. Then we propose a buffer mechanism for the first time to improve the robustness of the segmentation. During the optimization, we use global information to optimize the frames with large error, and interpolation instead of optimization to update well-estimated frames to hierarchically allocate the amount of computation according to error of each frame. Comparative experiments on the benchmark show that our method greatly improves the efficiency of optimization with almost no drop in accuracy, and outperforms existing high-efficiency optimization method by a large margin.
Training a small student network with the guidance of a larger teacher network is an effective way to promote the performance of the student. Despite the different types, the guided knowledge used to distill is always kept unchanged for different teacher and student pairs in previous knowledge distillation methods. However, we find that teacher and student models with different networks or trained from different initialization could have distinct feature representations among different channels. (e.g. the high activated channel for different categories). We name this incongruous representation of channels as teacher-student knowledge discrepancy in the distillation process. Ignoring the knowledge discrepancy problem of teacher and student models will make the learning of student from teacher more difficult. To solve this problem, in this paper, we propose a novel student-dependent distillation method, knowledge consistent distillation, which makes teacher's knowledge more consistent with the student and provides the best suitable knowledge to different student networks for distillation. Extensive experiments on different datasets (CIFAR100, ImageNet, COCO) and tasks (image classification, object detection) reveal the widely existing knowledge discrepancy problem between teachers and students and demonstrate the effectiveness of our proposed method. Our method is very flexible that can be easily combined with other state-of-the-art approaches.