Multi-modality fusion and multi-task learning are becoming trendy in 3D autonomous driving scenario, considering robust prediction and computation budget. However, naively extending the existing framework to the domain of multi-modality multi-task learning remains ineffective and even poisonous due to the notorious modality bias and task conflict. Previous works manually coordinate the learning framework with empirical knowledge, which may lead to sub-optima. To mitigate the issue, we propose a novel yet simple multi-level gradient calibration learning framework across tasks and modalities during optimization. Specifically, the gradients, produced by the task heads and used to update the shared backbone, will be calibrated at the backbone's last layer to alleviate the task conflict. Before the calibrated gradients are further propagated to the modality branches of the backbone, their magnitudes will be calibrated again to the same level, ensuring the downstream tasks pay balanced attention to different modalities. Experiments on large-scale benchmark nuScenes demonstrate the effectiveness of the proposed method, eg, an absolute 14.4% mIoU improvement on map segmentation and 1.4% mAP improvement on 3D detection, advancing the application of 3D autonomous driving in the domain of multi-modality fusion and multi-task learning. We also discuss the links between modalities and tasks.
相關內容
Learning effective recommendation models from sparse user interactions represents a fundamental challenge in developing sequential recommendation methods. Recently, pre-training-based methods have been developed to tackle this challenge. Though promising, in this paper, we show that existing methods suffer from the notorious negative transfer issue, where the model adapted from the pre-trained model results in worse performance compared to the model learned from scratch in the task of interest (i.e., target task). To address this issue, we develop a method, denoted as ANT, for transferable sequential recommendation. ANT mitigates negative transfer by 1) incorporating multi-modality item information, including item texts, images and prices, to effectively learn more transferable knowledge from related tasks (i.e., auxiliary tasks); and 2) better capturing task-specific knowledge in the target task using a re-learning-based adaptation strategy. We evaluate ANT against eight state-of-the-art baseline methods on five target tasks. Our experimental results demonstrate that ANT does not suffer from the negative transfer issue on any of the target tasks. The results also demonstrate that ANT substantially outperforms baseline methods in the target tasks with an improvement of as much as 15.2%. Our analysis highlights the superior effectiveness of our re-learning-based strategy compared to fine-tuning on the target tasks.
In computational pathology, multiple instance learning (MIL) is widely used to circumvent the computational impasse in giga-pixel whole slide image (WSI) analysis. It usually consists of two stages: patch-level feature extraction and slide-level aggregation. Recently, pretrained models or self-supervised learning have been used to extract patch features, but they suffer from low effectiveness or inefficiency due to overlooking the task-specific supervision provided by slide labels. Here we propose a weakly-supervised Label-Efficient WSI Screening method, dubbed LESS, for cytological WSI analysis with only slide-level labels, which can be effectively applied to small datasets. First, we suggest using variational positive-unlabeled (VPU) learning to uncover hidden labels of both benign and malignant patches. We provide appropriate supervision by using slide-level labels to improve the learning of patch-level features. Next, we take into account the sparse and random arrangement of cells in cytological WSIs. To address this, we propose a strategy to crop patches at multiple scales and utilize a cross-attention vision transformer (CrossViT) to combine information from different scales for WSI classification. The combination of our two steps achieves task-alignment, improving effectiveness and efficiency. We validate the proposed label-efficient method on a urine cytology WSI dataset encompassing 130 samples (13,000 patches) and FNAC 2019 dataset with 212 samples (21,200 patches). The experiment shows that the proposed LESS reaches 84.79%, 85.43%, 91.79% and 78.30% on a urine cytology WSI dataset, and 96.88%, 96.86%, 98.95%, 97.06% on FNAC 2019 dataset in terms of accuracy, AUC, sensitivity and specificity. It outperforms state-of-the-art MIL methods on pathology WSIs and realizes automatic cytological WSI cancer screening.
Audio-visual representation learning aims to develop systems with human-like perception by utilizing correlation between auditory and visual information. However, current models often focus on a limited set of tasks, and generalization abilities of learned representations are unclear. To this end, we propose the AV-SUPERB benchmark that enables general-purpose evaluation of unimodal audio/visual and bimodal fusion representations on 7 datasets covering 5 audio-visual tasks in speech and audio processing. We evaluate 5 recent self-supervised models and show that none of these models generalize to all tasks, emphasizing the need for future study on improving universal model performance. In addition, we show that representations may be improved with intermediate-task fine-tuning and audio event classification with AudioSet serves as a strong intermediate task. We release our benchmark with evaluation code and a model submission platform to encourage further research in audio-visual learning.
Over-the-air federated learning (OTA-FL) integrates communication and model aggregation by exploiting the innate superposition property of wireless channels. The approach renders bandwidth efficient learning, but requires care in handling the wireless physical layer impairments. In this paper, federated edge learning is considered for a network that is heterogeneous with respect to client (edge node) data set distributions and individual client resources, under a general non-convex learning objective. We augment the wireless OTA-FL system with a Reconfigurable Intelligent Surface (RIS) to enable a propagation environment with improved learning performance in a realistic time varying physical layer. Our approach is a cross-layer perspective that jointly optimizes communication, computation and learning resources, in this general heterogeneous setting. We adapt the local computation steps and transmission power of the clients in conjunction with the RIS phase shifts. The resulting joint communication and learning algorithm, RIS-assisted Over-the-air Adaptive Resource Allocation for Federated learning (ROAR-Fed) is shown to be convergent in this general setting. Numerical results demonstrate the effectiveness of ROAR-Fed under heterogeneous (non i.i.d.) data and imperfect CSI, indicating the advantage of RIS assisted learning in this general set up.
World models, especially in autonomous driving, are trending and drawing extensive attention due to their capacity for comprehending driving environments. The established world model holds immense potential for the generation of high-quality driving videos, and driving policies for safe maneuvering. However, a critical limitation in relevant research lies in its predominant focus on gaming environments or simulated settings, thereby lacking the representation of real-world driving scenarios. Therefore, we introduce DriveDreamer, a pioneering world model entirely derived from real-world driving scenarios. Regarding that modeling the world in intricate driving scenes entails an overwhelming search space, we propose harnessing the powerful diffusion model to construct a comprehensive representation of the complex environment. Furthermore, we introduce a two-stage training pipeline. In the initial phase, DriveDreamer acquires a deep understanding of structured traffic constraints, while the subsequent stage equips it with the ability to anticipate future states. The proposed DriveDreamer is the first world model established from real-world driving scenarios. We instantiate DriveDreamer on the challenging nuScenes benchmark, and extensive experiments verify that DriveDreamer empowers precise, controllable video generation that faithfully captures the structural constraints of real-world traffic scenarios. Additionally, DriveDreamer enables the generation of realistic and reasonable driving policies, opening avenues for interaction and practical applications.
Reinforcement learning-based policies for continuous control robotic navigation tasks often fail to adapt to changes in the environment during real-time deployment, which may result in catastrophic failures. To address this limitation, we propose a novel approach called RE-MOVE (REquest help and MOVE on) to adapt already trained policy to real-time changes in the environment without re-training via utilizing a language-based feedback. The proposed approach essentially boils down to addressing two main challenges of (1) when to ask for feedback and, if received, (2) how to incorporate feedback into trained policies. RE-MOVE incorporates an epistemic uncertainty-based framework to determine the optimal time to request instructions-based feedback. For the second challenge, we employ a zero-shot learning natural language processing (NLP) paradigm with efficient, prompt design and leverage state-of-the-art GPT-3.5, Llama-2 language models. To show the efficacy of the proposed approach, we performed extensive synthetic and real-world evaluations in several test-time dynamic navigation scenarios. Utilizing RE-MOVE result in up to 80% enhancement in the attainment of successful goals, coupled with a reduction of 13.50% in the normalized trajectory length, as compared to alternative approaches, particularly in demanding real-world environments with perceptual challenges.
Achieving accurate, efficient, and consistent localization within an a priori environment map remains a fundamental challenge in robotics and computer vision. Conventional map-based keyframe localization often suffers from sub-optimal viewpoints due to limited field of view (FOV), thus degrading its performance. To address this issue, in this paper, we design a real-time tightly-coupled Neural Radiance Fields (NeRF)-aided visual-inertial navigation system (VINS), termed NeRF-VINS. By effectively leveraging NeRF's potential to synthesize novel views, essential for addressing limited viewpoints, the proposed NeRF-VINS optimally fuses IMU and monocular image measurements along with synthetically rendered images within an efficient filter-based framework. This tightly coupled integration enables 3D motion tracking with bounded error. We extensively compare the proposed NeRF-VINS against the state-of-the-art methods that use prior map information, which is shown to achieve superior performance. We also demonstrate the proposed method is able to perform real-time estimation at 15 Hz, on a resource-constrained Jetson AGX Orin embedded platform with impressive accuracy.
The parallel alternating direction method of multipliers (ADMM) algorithms have gained popularity in statistics and machine learning for their efficient handling of large sample data problems. However, the parallel structure of these algorithms is based on the consensus problem, which can lead to an excessive number of auxiliary variables for high-dimensional data. In this paper, we propose a partition-insensitive parallel framework based on the linearized ADMM (LADMM) algorithm and apply it to solve nonconvex penalized smooth quantile regression problems. Compared to existing parallel ADMM algorithms, our algorithm does not rely on the consensus problem, resulting in a significant reduction in the number of variables that need to be updated at each iteration. It is worth noting that the solution of our algorithm remains unchanged regardless of how the total sample is divided, which is also known as partition-insensitivity. Furthermore, under some mild assumptions, we prove that the iterative sequence generated by the parallel LADMM algorithm converges to a critical point of the nonconvex optimization problem. Numerical experiments on synthetic and real datasets demonstrate the feasibility and validity of the proposed algorithm.
Spatio-temporal representation learning is critical for video self-supervised representation. Recent approaches mainly use contrastive learning and pretext tasks. However, these approaches learn representation by discriminating sampled instances via feature similarity in the latent space while ignoring the intermediate state of the learned representations, which limits the overall performance. In this work, taking into account the degree of similarity of sampled instances as the intermediate state, we propose a novel pretext task - spatio-temporal overlap rate (STOR) prediction. It stems from the observation that humans are capable of discriminating the overlap rates of videos in space and time. This task encourages the model to discriminate the STOR of two generated samples to learn the representations. Moreover, we employ a joint optimization combining pretext tasks with contrastive learning to further enhance the spatio-temporal representation learning. We also study the mutual influence of each component in the proposed scheme. Extensive experiments demonstrate that our proposed STOR task can favor both contrastive learning and pretext tasks. The joint optimization scheme can significantly improve the spatio-temporal representation in video understanding. The code is available at //github.com/Katou2/CSTP.
Network representation learning in low dimensional vector space has attracted considerable attention in both academic and industrial domains. Most real-world networks are dynamic with addition/deletion of nodes and edges. The existing graph embedding methods are designed for static networks and they cannot capture evolving patterns in a large dynamic network. In this paper, we propose a dynamic embedding method, dynnode2vec, based on the well-known graph embedding method node2vec. Node2vec is a random walk based embedding method for static networks. Applying static network embedding in dynamic settings has two crucial problems: 1) Generating random walks for every time step is time consuming 2) Embedding vector spaces in each timestamp are different. In order to tackle these challenges, dynnode2vec uses evolving random walks and initializes the current graph embedding with previous embedding vectors. We demonstrate the advantages of the proposed dynamic network embedding by conducting empirical evaluations on several large dynamic network datasets.
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191