arXiv subjects

Yue Wang

Publications and source records attributed to Yue Wang.

At least 739 records · Page 41Linked to original sources

arXiv2021

Joint Optimization of Communications and Federated Learning Over the Air

Federated learning (FL) is an attractive paradigm for making use of rich distributed data while protecting data privacy. Nonetheless, nonideal communication links and limited transmission resources may hinder the implementation of fast and accurate FL. In this paper, we study joint optimization of communications and FL based on analog aggregation transmission in realistic wireless networks. We first derive closed-form expressions for the expected convergence rate of FL over the air, which theoretically quantify the impact of analog aggregation on FL. Based on the analytical results, we develop a joint optimization model for accurate FL implementation, which allows a parameter server to select a subset of workers and determine an appropriate power scaling factor. Since the practical setting of FL over the air encounters unobservable parameters, we reformulate the joint optimization of worker selection and power allocation using controlled approximation. Finally, we efficiently solve the resulting mixed-integer programming problem via a simple yet optimal finite-set search method by reducing the search space. Simulation results show that the proposed solutions developed for realistic wireless analog channels outperform a benchmark method, and achieve comparable performance of the ideal case where FL is implemented over noise-free wireless channels.

cs.LG↗

arXiv2021

Learning Stereopsis from Geometric Synthesis for 6D Object Pose Estimation

Current monocular-based 6D object pose estimation methods generally achieve less competitive results than RGBD-based methods, mostly due to the lack of 3D information. To make up this gap, this paper proposes a 3D geometric volume based pose estimation method with a short baseline two-view setting. By constructing a geometric volume in the 3D space, we combine the features from two adjacent images to the same 3D space. Then a network is trained to learn the distribution of the position of object keypoints in the volume, and a robust soft RANSAC solver is deployed to solve the pose in closed form. To balance accuracy and cost, we propose a coarse-to-fine framework to improve the performance in an iterative way. The experiments show that our method outperforms state-of-the-art monocular-based methods, and is robust in different objects and scenes, especially in serious occlusion situations.

cs.CV↗

arXiv2021

Domain Generalization for Vision-based Driving Trajectory Generation

One of the challenges in vision-based driving trajectory generation is dealing with out-of-distribution scenarios. In this paper, we propose a domain generalization method for vision-based driving trajectory generation for autonomous vehicles in urban environments, which can be seen as a solution to extend the Invariant Risk Minimization (IRM) method in complex problems. We leverage an adversarial learning approach to train a trajectory generator as the decoder. Based on the pre-trained decoder, we infer the latent variables corresponding to the trajectories, and pre-train the encoder by regressing the inferred latent variable. Finally, we fix the decoder but fine-tune the encoder with the final trajectory loss. We compare our proposed method with the state-of-the-art trajectory generation method and some recent domain generalization methods on both datasets and simulation, demonstrating that our method has better generalization ability.

cs.CV↗

arXiv2021

HiTMap: A Hierarchical Topological Map Representation for Navigation in Unknown Environments

The ability to autonomously navigate in unknown environments is important for mobile robots. The map is the core component to achieve this. Most map representations rely on drift-free state estimation and provide a global metric map to navigate. However, in large-scale real-world applications, it's hard to prohibit drifts and compose a globally consistent map quickly. In this paper, a novel representation named, HiTMap, is proposed to enhance the existing map representations. The central idea is to adopt a submap-based hierarchical topology rather than a global metric map so that only a local metric map is maintained for obstacle avoidance which ensures the lightweight of the representation. To guide the robots navigate into unknown spaces, frontiers are detected and attached to the map as an attribute. We also develop a path planning module to evaluate the feasibility and efficiency of our map representation. The system is validated in a simulation environment and a demonstration in the real world is conducted. In addition, the HiTMap is made available open-source.

cs.RO↗

arXiv2021

Nodal profile control for networks of geometrically exact beams

In this work, we consider networks of so-called geometrically exact beams, namely, shearable beams that may undergo large motions. The corresponding mathematical model, commonly written in terms of displacements and rotations expressed in a fixed basis (Geometrically Exact Beam model, or GEB), has a quasilinear governing system. However, the model may also be written in terms of intrinsic variables expressed in a moving basis attached to the beam (Intrinsic GEB model, or IGEB) and while the number of equations is then doubled, the latter model has the advantage of being of first-order, hyperbolic and only semilinear. First, for any network, we show the existence and uniqueness of semi-global in time classical solutions to the IGEB model (i.e., for arbitrarily large time intervals, provided that the data are small enough). Then, for a specific network containing a cycle, we address the problem of local exact controllability of nodal profiles for the IGEB model -- we steer the solution to satisfy given profiles at one of the multiple nodes by means of controls applied at the simple nodes -- by using the constructive method of Zhuang, Leugering and Li [Exact boundary controllability of nodal profile for Saint-Venant system on a network with loops, in J. Math. Pures Appl., 2018]. Afterwards, for any network, we show that the existence of a unique classical solution to the IGEB network implies the same for the corresponding GEB network, by using that these two models are related by a nonlinear transformation. In particular, this allows us to give corresponding existence, uniqueness and controllability results for the GEB network.

math.OC↗

arXiv2021

Learning Observation-Based Certifiable Safe Policy for Decentralized Multi-Robot Navigation

Safety is of great importance in multi-robot navigation problems. In this paper, we propose a control barrier function (CBF) based optimizer that ensures robot safety with both high probability and flexibility, using only sensor measurement. The optimizer takes action commands from the policy network as initial values and then provides refinement to drive the potentially dangerous ones back into safe regions. With the help of a deep transition model that predicts the evolution of surrounding dynamics and the consequences of different actions, the CBF module can guide the optimization in a reasonable time horizon. We also present a novel joint training framework that improves the cooperation between the Reinforcement Learning (RL) based policy and the CBF-based optimizer both in training and inference procedures by utilizing reward feedback from the CBF module. We observe that the policy using our method can achieve a higher success rate while maintaining the safety of multiple robots in significantly fewer episodes compared with other methods. Experiments are conducted in multiple scenarios both in simulation and the real world, the results demonstrate the effectiveness of our method in maintaining the safety of multi-robot navigation. Code is available at \url{https://github.com/YuxiangCui/MARL-OCBF

cs.RO↗

arXiv2021

A Right Invariant Extended Kalman Filter for Object based SLAM

With the recent advance of deep learning based object recognition and estimation, it is possible to consider object level SLAM where the pose of each object is estimated in the SLAM process. In this paper, based on a novel Lie group structure, a right invariant extended Kalman filter (RI-EKF) for object based SLAM is proposed. The observability analysis shows that the proposed algorithm automatically maintains the correct unobservable subspace, while standard EKF (Std-EKF) based SLAM algorithm does not. This results in a better consistency for the proposed algorithm comparing to Std-EKF. Finally, simulations and real world experiments validate not only the consistency and accuracy of the proposed algorithm, but also the practicability of the proposed RI-EKF for object based SLAM problem. The MATLAB code of the algorithm is made publicly available.

cs.RO↗

arXiv2021

Asymptotic Behavior of the Steady Prandtl Equation

We study the asymptotic behavior of the Oleinik's solution to the steady Prandtl equation when the outer flow $U(x)=1$. Serrin proved that the Oleinik's solution converges to the famous Blasius solution $\bar u$ in $L^\infty_y$ sense as $x\rightarrow+\infty$. The explicit decay estimates of $u-\bar u$ and its derivatives were proved by Iyer[ARMA 237(2020)] when the initial data is a small localized perturbation of the Blasius profile. In this paper, we prove the explicit decay estimate of $\|u(x,y)-\bar{u}(x,y)\|_{L^\infty_y}$ for general initial data with exponential decay. We also prove the decay estimates of its derivatives when the data has an additional concave assumption. Our proof is based on the maximum principle technique. The key ingredient is to find a series of barrier functions.

math.AP↗

arXiv2021

Socially-Aware Multi-Agent Following with 2D Laser Scans via Deep Reinforcement Learning and Potential Field

Target following in dynamic pedestrian environments is an important task for mobile robots. However, it is challenging to keep tracking the target while avoiding collisions in crowded environments, especially with only one robot. In this paper, we propose a multi-agent method for an arbitrary number of robots to follow the target in a socially-aware manner using only 2D laser scans. The multi-agent following problem is tackled by utilizing the complementary strengths of both reinforcement learning and potential field, in which the reinforcement learning part handles local interactions while navigating to the goals assigned by the potential field. Specifically, with the help of laser scans in obstacle map representation, the learning-based policy can help the robots avoid collisions with both static obstacles and dynamic obstacles like pedestrians in advance, namely socially aware. While the formation control and goal assignment for each robot is obtained from a target-centered potential field constructed using aggregated state information from all the following robots. Experiments are conducted in multiple settings, including random obstacle distributions and different numbers of robots. Results show that our method works successfully in unseen dynamic environments. The robots can follow the target in a socially compliant manner with only 2D laser scans.

cs.RO↗

arXiv2021

CodeT5: Identifier-aware Unified Pre-trained Encoder-Decoder Models for Code Understanding and Generation

Pre-trained models for Natural Languages (NL) like BERT and GPT have been recently shown to transfer well to Programming Languages (PL) and largely benefit a broad set of code-related tasks. Despite their success, most current methods either rely on an encoder-only (or decoder-only) pre-training that is suboptimal for generation (resp. understanding) tasks or process the code snippet in the same way as NL, neglecting the special characteristics of PL such as token types. We present CodeT5, a unified pre-trained encoder-decoder Transformer model that better leverages the code semantics conveyed from the developer-assigned identifiers. Our model employs a unified framework to seamlessly support both code understanding and generation tasks and allows for multi-task learning. Besides, we propose a novel identifier-aware pre-training task that enables the model to distinguish which code tokens are identifiers and to recover them when they are masked. Furthermore, we propose to exploit the user-written code comments with a bimodal dual generation task for better NL-PL alignment. Comprehensive experiments show that CodeT5 significantly outperforms prior methods on understanding tasks such as code defect detection and clone detection, and generation tasks across various directions including PL-NL, NL-PL, and PL-PL. Further analysis reveals that our model can better capture semantic information from code. Our code and pre-trained models are released at https: //github.com/salesforce/CodeT5 .

cs.CL↗

arXiv2021

Stiffened gas approximation and GRP resolution for fluid flows of real materials

The equation of state (EOS) embodies thermodynamic properties of compressible fluid materials and usually has very complicated forms in real engineering applications, subject to the physical requirements of thermodynamics. The complexity of EOS in form gives rise to the difficulty in analyzing relevant wave patterns. Concerning the design of numerical algorithms, the complex EOS causes the inefficiency of Riemann solvers and even the loss of robustness, which hampers the development of Godunov-type numerical schemes. In this paper, a strategy of local stiffened gas approximation is proposed for real materials. The stiffened gas EOS is used to approximate general EOS locally at each interface of computational control volumes so that the Riemann solver can be significantly simplified. In the meantime, the generalized Riemann problem (GRP) solver is adopted not only for high resolution purpose but effective reflection of the local thermodynamics as well. The resulting scheme is demonstrated to be efficient and robust and numerical examples display the excellent performance of such an approximation.

math.NA↗

arXiv2021

A framework for massive scale personalized promotion

Technology companies building consumer-facing platforms may have access to massive-scale user population. In recent years, promotion with quantifiable incentive has become a popular approach for increasing active users on such platforms. On one hand, increased user activities can introduce network effect, bring in advertisement audience, and produce other benefits. On the other hand, massive-scale promotion causes massive cost. Therefore making promotion campaigns efficient in terms of return-on-investment (ROI) is of great interest to many companies. This paper proposes a practical two-stage framework that can optimize the ROI of various massive-scale promotion campaigns. In the first stage, users' personal promotion-response curves are modeled by machine learning techniques. In the second stage, business objectives and resource constraints are formulated into an optimization problem, the decision variables of which are how much incentive to give to each user. In order to do effective optimization in the second stage, counterfactual prediction and noise-reduction are essential for the first stage. We leverage existing counterfactual prediction techniques to correct treatment bias in data. We also introduce a novel deep neural network (DNN) architecture, the deep-isotonic-promotion-network (DIPN), to reduce noise in the promotion response curves. The DIPN architecture incorporates our prior knowledge of response curve shape, by enforcing isotonicity and smoothness. It out-performed regular DNN and other state-of-the-art shape-constrained models in our experiments.

cs.LG↗

arXiv2021

Stop-and-Go: Exploring Backdoor Attacks on Deep Reinforcement Learning-based Traffic Congestion Control Systems

Recent work has shown that the introduction of autonomous vehicles (AVs) in traffic could help reduce traffic jams. Deep reinforcement learning methods demonstrate good performance in complex control problems, including autonomous vehicle control, and have been used in state-of-the-art AV controllers. However, deep neural networks (DNNs) render automated driving vulnerable to machine learning-based attacks. In this work, we explore the backdooring/trojanning of DRL-based AV controllers. We develop a trigger design methodology that is based on well-established principles of traffic physics. The malicious actions include vehicle deceleration and acceleration to cause stop-and-go traffic waves to emerge (congestion attacks) or AV acceleration resulting in the AV crashing into the vehicle in front (insurance attack). We test our attack on single-lane and two-lane circuits. Our experimental results show that the backdoored model does not compromise normal operation performance, with the maximum decrease in cumulative rewards being 1%. Still, it can be maliciously activated to cause a crash or congestion when the corresponding triggers appear.

cs.CR↗

arXiv2021

On Feature Decorrelation in Self-Supervised Learning

In self-supervised representation learning, a common idea behind most of the state-of-the-art approaches is to enforce the robustness of the representations to predefined augmentations. A potential issue of this idea is the existence of completely collapsed solutions (i.e., constant features), which are typically avoided implicitly by carefully chosen implementation details. In this work, we study a relatively concise framework containing the most common components from recent approaches. We verify the existence of complete collapse and discover another reachable collapse pattern that is usually overlooked, namely dimensional collapse. We connect dimensional collapse with strong correlations between axes and consider such connection as a strong motivation for feature decorrelation (i.e., standardizing the covariance matrix). The gains from feature decorrelation are verified empirically to highlight the importance and the potential of this insight.

cs.LG↗

arXiv2021

A Multi-Hypothesis Approach to Pose Ambiguity in Object-Based SLAM

In object-based Simultaneous Localization and Mapping (SLAM), 6D object poses offer a compact representation of landmark geometry useful for downstream planning and manipulation tasks. However, measurement ambiguity then arises as objects may possess complete or partial object shape symmetries (e.g., due to occlusion), making it difficult or impossible to generate a single consistent object pose estimate. One idea is to generate multiple pose candidates to counteract measurement ambiguity. In this paper, we develop a novel approach that enables an object-based SLAM system to reason about multiple pose hypotheses for an object, and synthesize this locally ambiguous information into a globally consistent robot and landmark pose estimation formulation. In particular, we (1) present a learned pose estimation network that provides multiple hypotheses about the 6D pose of an object; (2) by treating the output of our network as components of a mixture model, we incorporate pose predictions into a SLAM system, which, over successive observations, recovers a globally consistent set of robot and object (landmark) pose estimates. We evaluate our approach on the popular YCB-Video Dataset and a simulated video featuring YCB objects. Experiments demonstrate that our approach is effective in improving the robustness of object-based SLAM in the face of object pose ambiguity.

cs.RO↗

arXiv2021

CORAL: Colored structural representation for bi-modal place recognition

Place recognition is indispensable for a drift-free localization system. Due to the variations of the environment, place recognition using single-modality has limitations. In this paper, we propose a bi-modal place recognition method, which can extract a compound global descriptor from the two modalities, vision and LiDAR. Specifically, we first build the elevation image generated from 3D points as a structural representation. Then, we derive the correspondences between 3D points and image pixels that are further used in merging the pixel-wise visual features into the elevation map grids. In this way, we fuse the structural features and visual features in the consistent bird-eye view frame, yielding a semantic representation, namely CORAL. And the whole network is called CORAL-VLAD. Comparisons on the Oxford RobotCar show that CORAL-VLAD has superior performance against other state-of-the-art methods. We also demonstrate that our network can be generalized to other scenes and sensor configurations on cross-city datasets.

cs.CV↗

arXiv2021

COKE: Communication-Censored Decentralized Kernel Learning

This paper studies the decentralized optimization and learning problem where multiple interconnected agents aim to learn an optimal decision function defined over a reproducing kernel Hilbert space by jointly minimizing a global objective function, with access to their own locally observed dataset. As a non-parametric approach, kernel learning faces a major challenge in distributed implementation: the decision variables of local objective functions are data-dependent and thus cannot be optimized under the decentralized consensus framework without any raw data exchange among agents. To circumvent this major challenge, we leverage the random feature (RF) approximation approach to enable consensus on the function modeled in the RF space by data-independent parameters across different agents. We then design an iterative algorithm, termed DKLA, for fast-convergent implementation via ADMM. Based on DKLA, we further develop a communication-censored kernel learning (COKE) algorithm that reduces the communication load of DKLA by preventing an agent from transmitting at every iteration unless its local updates are deemed informative. Theoretical results in terms of linear convergence guarantee and generalization performance analysis of DKLA and COKE are provided. Comprehensive tests on both synthetic and real datasets are conducted to verify the communication efficiency and learning effectiveness of COKE.

cs.LG↗

arXiv2021

Efficient learning of goal-oriented push-grasping synergy in clutter

We focus on the task of goal-oriented grasping, in which a robot is supposed to grasp a pre-assigned goal object in clutter and needs some pre-grasp actions such as pushes to enable stable grasps. However, in this task, the robot gets positive rewards from environment only when successfully grasping the goal object. Besides, joint pushing and grasping elongates the action sequence, compounding the problem of reward delay. Thus, sample inefficiency remains a main challenge in this task. In this paper, a goal-conditioned hierarchical reinforcement learning formulation with high sample efficiency is proposed to learn a push-grasping policy for grasping a specific object in clutter. In our work, sample efficiency is improved by two means. First, we use a goal-conditioned mechanism by goal relabeling to enrich the replay buffer. Second, the pushing and grasping policies are respectively regarded as a generator and a discriminator and the pushing policy is trained with supervision of the grasping discriminator, thus densifying pushing rewards. To deal with the problem of distribution mismatch caused by different training settings of two policies, an alternating training stage is added to learn pushing and grasping in turn. A series of experiments carried out in simulation and real world indicate that our method can quickly learn effective pushing and grasping policies and outperforms existing methods in task completion rate and goal grasp success rate by less times of motion. Furthermore, we validate that our system can also adapt to goal-agnostic conditions with better performance. Note that our system can be transferred to the real world without any fine-tuning. Our code is available at https://github.com/xukechun/Efficient_goal-oriented_push-grasping_synergy.

cs.RO↗