Humans can quickly learn new concepts with limited experience, while not forgetting learned knowledge. Such ability in machine learning is referred to as few-shot class-incremental learning (FSCIL). Although some methods try to solve this problem by putting similar efforts to prevent forgetting and promote learning, we find existing techniques do not give enough importance to the new category as new training samples are rather rare. In this article, we propose a new biased-to-unbiased rectification method, which introduces a trainable transition matrix to mitigate the prediction discrepancy between the old classes and the new classes. This transition matrix is to be diagonally dominated, normalized, and differentiable with new-knowledge-preferred prior, to solving the strong bias between heavy old knowledge and limited new knowledge. Hence, we can achieve a balanced solution between learning new concepts and preventing catastrophic forgetting by giving new classes more chances. Extensive experiments on miniImagenet, CIFAR100, and CUB200 demonstrate that our method outperforms the latest state-of-the-art methods by 1.1%, 1.44%, and 2.08%, respectively.

With the increase of diversity in application preferences and networks, existing congestion control algorithms (CCAs) do not accommodate this complicated reality. Previous classic CCAs are designed for a specific domain with fixed rules, failing to adapt to such diversities. Recently surged learning-based CCAs have great potential in adaptability and flexibility but are not practical due to unsatisfying performance on convergence, fairness, overhead, consistency and safety assurance. In this paper, we propose Libra, a unified congestion control framework, that can empower these properties by combining the wisdom of classic and reinforcement learning (RL)-based CCAs. Extensive evaluation of Libra’s Linux kernel implementations on both live Internet and emulated networks shows performance improvement under dynamic networks (e.g.,$1.2\times $throughput than Orca on average). At the same time, Libra can flexibly satisfy different application needs, reduce the running overhead by at most$0.88\times $and perform good fairness and convergence properties, well-fitting our theoretical analysis.

Class-incremental object detection (CIOD) is a real-world desired capability, requiring an object detector to continuously adapt to new tasks without forgetting learned ones, with the main challenge being catastrophic forgetting. Many methods based on distillation and replay have been proposed to alleviate this problem. However, they typically learn on a pure visual backbone, neglecting the powerful representation capabilities of textual cues, which to some extent limits their performance. In this paper, we propose task-aware language-image representation to mitigate catastrophic forgetting, introducing a new paradigm for language-image-based CIOD. First of all, we demonstrate the significant advantage of language-image detectors in mitigating catastrophic forgetting. Secondly, we propose a learning task-aware language-image representation method that overcomes the existing drawback of directly utilizing the language-image detector for CIOD. More specifically, we learn the language-image representation of different tasks through an insulating approach in the training stage, while using the alignment scores produced by task-specific language-image representation in the inference stage. Through our proposed method, language-image detectors can be more practical for CIOD. We conduct extensive experiments on COCO 2017 and Pascal VOC 2007 and demonstrate that the proposed method achieves state-of-the-art results under the various CIOD settings.

Ride-hailing service (RHS) has become an important transportation mode in our daily life. Although many works have been proposed to improve RHS from different aspects, only few works focus on the selections of pick-up locations, where rider and driver meet and start a trip. In this paper, we presentMPLRec, a data-driven pick-up location recommendation system that exploits riders' specific mobility demands,e.g.destination, and historical experiences to meet riders' travel requirements.MPLRecgenerates potential pick-up locations over the road network and characterizes them with rich features that describe a location from the riders' perspective. We also build spatio-temporal indexes to organize potential pick-up locations and historical data for facilitating online recommending. When processing an online recommendation request,MPLRecderives candidate pick-up locations and investigates them with materialized features, which are computed from historical order and trajectory data while considering rider's mobility demands. Based on these features, a novel scoring function is used to derive the best pick-up location for each request. Moreover, we implement an RHS simulator to evaluateMPLRecusing large-scale practical ride-hailing datasets. Extensive experiments and simulations demonstrate the effectiveness and efficiency ofMPLRec, which can complete each request within 0.5 s and largely reduce the ride-hailing costs when compared to baseline methods.