In cooperative multi-agent tasks, a team of agents jointly interact with an environment by taking actions, receiving a team reward and observing the next state. During the interactions, the uncertainty of environment and reward will inevitably induce stochasticity in the long-term returns and the randomness can be exacerbated with the increasing number of agents. However, most of the existing value-based multi-agent reinforcement learning (MARL) methods only model the expectations of individual Q-values and global Q-value, ignoring such randomness. Compared to the expectations of the long-term returns, it is more preferable to directly model the stochasticity by estimating the returns through distributions. With this motivation, this work proposes DQMIX, a novel value-based MARL method, from a distributional perspective. Specifically, we model each individual Q-value with a categorical distribution. To integrate these individual Q-value distributions into the global Q-value distribution, we design a distribution mixing network, based on five basic operations on the distribution. We further prove that DQMIX satisfies the \emph{Distributional-Individual-Global-Max} (DIGM) principle with respect to the expectation of distribution, which guarantees the consistency between joint and individual greedy action selections in the global Q-value and individual Q-values. To validate DQMIX, we demonstrate its ability to factorize a matrix game with stochastic rewards. Furthermore, the experimental results on a challenging set of StarCraft II micromanagement tasks show that DQMIX consistently outperforms the value-based multi-agent reinforcement learning baselines.
In cooperative multi-agent systems, agents jointly take actions and receive a team reward instead of individual rewards. In the absence of individual reward signals, credit assignment mechanisms are usually introduced to discriminate the contributions of different agents so as to achieve effective cooperation. Recently, the value decomposition paradigm has been widely adopted to realize credit assignment, and QMIX has become the state-of-the-art solution. In this paper, we revisit QMIX from two aspects. First, we propose a new perspective on credit assignment measurement and empirically show that QMIX suffers limited discriminability on the assignment of credits to agents. Second, we propose a gradient entropy regularization with QMIX to realize a discriminative credit assignment, thereby improving the overall performance. The experiments demonstrate that our approach can comparatively improve learning efficiency and achieve better performance.
Molecular conformation generation aims to generate three-dimensional coordinates of all the atoms in a molecule and is an important task in bioinformatics and pharmacology. Previous distance-based methods first predict interatomic distances and then generate conformations based on them, which could result in conflicting distances. In this work, we propose a method that directly predicts the coordinates of atoms. We design a dedicated loss function for conformation generation, which is invariant to roto-translation of coordinates of conformations and permutation of symmetric atoms in molecules. We further design a backbone model that stacks multiple blocks, where each block refines the conformation generated by its preceding block. Our method achieves state-of-the-art results on four public benchmarks: on small-scale GEOM-QM9 and GEOM-Drugs which have $200$K training data, we can improve the previous best matching score by $3.5\%$ and $28.9\%$; on large-scale GEOM-QM9 and GEOM-Drugs which have millions of training data, those two improvements are $47.1\%$ and $36.3\%$. This shows the effectiveness of our method and the great potential of the direct approach. Our code is released at \url{https://github.com/DirectMolecularConfGen/DMCG}.
Human actions are typically of combinatorial structures or patterns, i.e., subjects, objects, plus spatio-temporal interactions in between. Discovering such structures is therefore a rewarding way to reason about the dynamics of interactions and recognize the actions. In this paper, we introduce a new design of sub-graphs to represent and encode the discriminative patterns of each action in the videos. Specifically, we present MUlti-scale Sub-graph LEarning (MUSLE) framework that novelly builds space-time graphs and clusters the graphs into compact sub-graphs on each scale with respect to the number of nodes. Technically, MUSLE produces 3D bounding boxes, i.e., tubelets, in each video clip, as graph nodes and takes dense connectivity as graph edges between tubelets. For each action category, we execute online clustering to decompose the graph into sub-graphs on each scale through learning Gaussian Mixture Layer and select the discriminative sub-graphs as action prototypes for recognition. Extensive experiments are conducted on both Something-Something V1 & V2 and Kinetics-400 datasets, and superior results are reported when comparing to state-of-the-art methods. More remarkably, our MUSLE achieves to-date the best reported accuracy of 65.0% on Something-Something V2 validation set.
Deep reinforcement learning algorithms can perform poorly in real-world tasks due to the discrepancy between source and target environments. This discrepancy is commonly viewed as the disturbance in transition dynamics. Many existing algorithms learn robust policies by modeling the disturbance and applying it to source environments during training, which usually requires prior knowledge about the disturbance and control of simulators. However, these algorithms can fail in scenarios where the disturbance from target environments is unknown or is intractable to model in simulators. To tackle this problem, we propose a novel model-free actor-critic algorithm -- namely, state-conservative policy optimization (SCPO) -- to learn robust policies without modeling the disturbance in advance. Specifically, SCPO reduces the disturbance in transition dynamics to that in state space and then approximates it by a simple gradient-based regularizer. The appealing features of SCPO include that it is simple to implement and does not require additional knowledge about the disturbance or specially designed simulators. Experiments in several robot control tasks demonstrate that SCPO learns robust policies against the disturbance in transition dynamics.
Model-based reinforcement learning algorithms, which aim to learn a model of the environment to make decisions, are more sample efficient than their model-free counterparts. The sample efficiency of model-based approaches relies on whether the model can well approximate the environment. However, learning an accurate model is challenging, especially in complex and noisy environments. To tackle this problem, we propose the conservative model-based actor-critic (CMBAC), a novel approach that achieves high sample efficiency without the strong reliance on accurate learned models. Specifically, CMBAC learns multiple estimates of the Q-value function from a set of inaccurate models and uses the average of the bottom-k estimates -- a conservative estimate -- to optimize the policy. An appealing feature of CMBAC is that the conservative estimates effectively encourage the agent to avoid unreliable "promising actions" -- whose values are high in only a small fraction of the models. Experiments demonstrate that CMBAC significantly outperforms state-of-the-art approaches in terms of sample efficiency on several challenging tasks, and the proposed method is more robust than previous methods in noisy environments.
In this paper, we propose a radio-assisted human detection framework by incorporating radio information into the state-of-the-art detection methods, including anchor-based onestage detectors and two-stage detectors. We extract the radio localization and identifer information from the radio signals to assist the human detection, due to which the problem of false positives and false negatives can be greatly alleviated. For both detectors, we use the confidence score revision based on the radio localization to improve the detection performance. For two-stage detection methods, we propose to utilize the region proposals generated from radio localization rather than relying on region proposal network (RPN). Moreover, with the radio identifier information, a non-max suppression method with the radio localization constraint has also been proposed to further suppress the false detections and reduce miss detections. Experiments on the simulative Microsoft COCO dataset and Caltech pedestrian datasets show that the mean average precision (mAP) and the miss rate of the state-of-the-art detection methods can be improved with the aid of radio information. Finally, we conduct experiments in real-world scenarios to demonstrate the feasibility of our proposed method in practice.
In image retrieval, deep local features learned in a data-driven manner have been demonstrated effective to improve retrieval performance. To realize efficient retrieval on large image database, some approaches quantize deep local features with a large codebook and match images with aggregated match kernel. However, the complexity of these approaches is non-trivial with large memory footprint, which limits their capability to jointly perform feature learning and aggregation. To generate compact global representations while maintaining regional matching capability, we propose a unified framework to jointly learn local feature representation and aggregation. In our framework, we first extract deep local features using CNNs. Then, we design a tokenizer module to aggregate them into a few visual tokens, each corresponding to a specific visual pattern. This helps to remove background noise, and capture more discriminative regions in the image. Next, a refinement block is introduced to enhance the visual tokens with self-attention and cross-attention. Finally, different visual tokens are concatenated to generate a compact global representation. The whole framework is trained end-to-end with image-level labels. Extensive experiments are conducted to evaluate our approach, which outperforms the state-of-the-art methods on the Revisited Oxford and Paris datasets.
It has been well recognized that fusing the complementary information from depth-aware LiDAR point clouds and semantic-rich stereo images would benefit 3D object detection. Nevertheless, it is not trivial to explore the inherently unnatural interaction between sparse 3D points and dense 2D pixels. To ease this difficulty, the recent proposals generally project the 3D points onto the 2D image plane to sample the image data and then aggregate the data at the points. However, this approach often suffers from the mismatch between the resolution of point clouds and RGB images, leading to sub-optimal performance. Specifically, taking the sparse points as the multi-modal data aggregation locations causes severe information loss for high-resolution images, which in turn undermines the effectiveness of multi-sensor fusion. In this paper, we present VPFNet -- a new architecture that cleverly aligns and aggregates the point cloud and image data at the `virtual' points. Particularly, with their density lying between that of the 3D points and 2D pixels, the virtual points can nicely bridge the resolution gap between the two sensors, and thus preserve more information for processing. Moreover, we also investigate the data augmentation techniques that can be applied to both point clouds and RGB images, as the data augmentation has made non-negligible contribution towards 3D object detectors to date. We have conducted extensive experiments on KITTI dataset, and have observed good performance compared to the state-of-the-art methods. Remarkably, our VPFNet achieves 83.21\% moderate 3D AP and 91.86\% moderate BEV AP on the KITTI test set, ranking the 1st since May 21th, 2021. The network design also takes computation efficiency into consideration -- we can achieve a FPS of 15 on a single NVIDIA RTX 2080Ti GPU. The code will be made available for reproduction and further investigation.