arXiv subjects

Hao Su

Publications and source records attributed to Hao Su.

At least 235 records · Page 13Linked to original sources

arXiv2019

Mapping State Space using Landmarks for Universal Goal Reaching

An agent that has well understood the environment should be able to apply its skills for any given goals, leading to the fundamental problem of learning the Universal Value Function Approximator (UVFA). A UVFA learns to predict the cumulative rewards between all state-goal pairs. However, empirically, the value function for long-range goals is always hard to estimate and may consequently result in failed policy. This has presented challenges to the learning process and the capability of neural networks. We propose a method to address this issue in large MDPs with sparse rewards, in which exploration and routing across remote states are both extremely challenging. Our method explicitly models the environment in a hierarchical manner, with a high-level dynamic landmark-based map abstracting the visited state space, and a low-level value network to derive precise local decisions. We use farthest point sampling to select landmark states from past experience, which has improved exploration compared with simple uniform sampling. Experimentally we showed that our method enables the agent to reach long-range goals at the early training stage, and achieve better performance than standard RL algorithms for a number of challenging tasks.

cs.LG↗

arXiv2019

Magnetic structure and excitations of the topological semimetal YbMnBi$_2$

We investigated the magnetic structure and dynamics of YbMnBi$_2$, with elastic and inelastic neutron scattering, to shed light on the topological nature of the charge carriers in the antiferromagnetic phase. We confirm C-type antiferromagnetic ordering of the Mn spins below $T_{\rm N} = 290$ K, and determine that the spins point along the $c$-axis to within about $3^\circ$. The observed magnon spectrum can be described very well by the same effective spin Hamiltonian as was used previously to model the magnon spectrum of CaMnBi$_2$. Our results show conclusively that the creation of Weyl nodes in YbMnBi$_2$ by the time-reversal-symmetry breaking mechanism can be excluded in the bulk.

cond-mat.str-el↗

arXiv2019

Point-Based Multi-View Stereo Network

We introduce Point-MVSNet, a novel point-based deep framework for multi-view stereo (MVS). Distinct from existing cost volume approaches, our method directly processes the target scene as point clouds. More specifically, our method predicts the depth in a coarse-to-fine manner. We first generate a coarse depth map, convert it into a point cloud and refine the point cloud iteratively by estimating the residual between the depth of the current iteration and that of the ground truth. Our network leverages 3D geometry priors and 2D texture information jointly and effectively by fusing them into a feature-augmented point cloud, and processes the point cloud to estimate the 3D flow for each point. This point-based architecture allows higher accuracy, more computational efficiency and more flexibility than cost-volume-based counterparts. Experimental results show that our approach achieves a significant improvement in reconstruction quality compared with state-of-the-art methods on the DTU and the Tanks and Temples dataset. Our source code and trained models are available at https://github.com/callmeray/PointMVSNet .

cs.CV↗

arXiv2019

StructureNet: Hierarchical Graph Networks for 3D Shape Generation

The ability to generate novel, diverse, and realistic 3D shapes along with associated part semantics and structure is central to many applications requiring high-quality 3D assets or large volumes of realistic training data. A key challenge towards this goal is how to accommodate diverse shape variations, including both continuous deformations of parts as well as structural or discrete alterations which add to, remove from, or modify the shape constituents and compositional structure. Such object structure can typically be organized into a hierarchy of constituent object parts and relationships, represented as a hierarchy of n-ary graphs. We introduce StructureNet, a hierarchical graph network which (i) can directly encode shapes represented as such n-ary graphs; (ii) can be robustly trained on large and complex shape families; and (iii) can be used to generate a great diversity of realistic structured shape geometries. Technically, we accomplish this by drawing inspiration from recent advances in graph neural networks to propose an order-invariant encoding of n-ary graphs, considering jointly both part geometry and inter-part relations during network training. We extensively evaluate the quality of the learned latent spaces for various shape families and show significant advantages over baseline and competing methods. The learned latent spaces enable several structure-aware geometry processing applications, including shape generation and interpolation, shape editing, or shape structure discovery directly from un-annotated images, point clouds, or partial scans.

cs.GR↗

arXiv2019

Design and Control of a Quasi-Direct Drive Soft Exoskeleton for Knee Injury Prevention during Squatting

This paper presents design and control innovations of wearable robots that tackle two barriers to widespread adoption of powered exoskeletons, namely restriction of human movement and versatile control of wearable co-robot systems. First, the proposed quasi-direct drive actuation comprising of our customized high torque density motors and low ratio transmission mechanism significantly reduces the mass of the robot and produces high backdrivability. Second, we derive a biomechanics model-based control that generates biological torque profile for versatile control of both squat and stoop lifting assistance. The control algorithm detects lifting postures using compact inertial measurement unit (IMU) sensors to generate an assistive profile that is proportional to the biological torque produced from our model. Experimental results demonstrate that the robot exhibits low mechanical impedance (1.5 Nm resistive torque) when it is unpowered and 0.5 Nm resistive torque with zero-torque tracking control. Root mean square (RMS) error of torque tracking is less than 0.29 Nm (1.21% error of 24 Nm peak torque). Compared with squatting without the exoskeleton, the controller reduces 87.5%, 80% and 75% of the of three knee extensor muscles (average peak EMG of 3 healthy subjects) during squat with 50% of biological torque assistance.

cs.RO↗

arXiv2019

Spine-Inspired Continuum Soft Exoskeleton for Stoop Lifting Assistance

Back injuries are the most prevalent work-related musculoskeletal disorders and represent a major cause of disability. Although innovations in wearable robots aim to alleviate this hazard, the majority of existing exoskeletons are obtrusive because the rigid linkage design limits natural movement, thus causing ergonomic risk. Moreover, these existing systems are typically only suitable for one type of movement assistance, not ubiquitous for a wide variety of activities. To fill in this gap, this paper presents a new wearable robot design approach continuum soft exoskeleton. This spine-inspired wearable robot is unobtrusive and assists both squat and stoops while not impeding walking motion. To tackle the challenge of the unique anatomy of spine that is inappropriate to be simplified as a single degree of freedom joint, our robot is conformal to human anatomy and it can reduce multiple types of forces along the human spine such as the spinae muscle force, shear, and compression force of the lumbar vertebrae. We derived kinematics and kinetics models of this mechanism and established an analytical biomechanics model of human-robot interaction. Quantitative analysis of disc compression force, disc shear force and muscle force was performed in simulation. We further developed a virtual impedance control strategy to deliver force control and compensate hysteresis of Bowden cable transmission. The feasibility of the prototype was experimentally tested on three healthy subjects. The root mean square error of force tracking is 6.63 N (3.3 % of the 200N peak force) and it demonstrated that it can actively control the stiffness to the desired value. This continuum soft exoskeleton represents a feasible solution with the potential to reduce back pain for multiple activities and multiple forces along the human spine.

cs.RO↗

arXiv2019

Extending Adversarial Attacks and Defenses to Deep 3D Point Cloud Classifiers

3D object classification and segmentation using deep neural networks has been extremely successful. As the problem of identifying 3D objects has many safety-critical applications, the neural networks have to be robust against adversarial changes to the input data set. There is a growing body of research on generating human-imperceptible adversarial attacks and defenses against them in the 2D image classification domain. However, 3D objects have various differences with 2D images, and this specific domain has not been rigorously studied so far. We present a preliminary evaluation of adversarial attacks on deep 3D point cloud classifiers, namely PointNet and PointNet++, by evaluating both white-box and black-box adversarial attacks that were proposed for 2D images and extending those attacks to reduce the perceptibility of the perturbations in 3D space. We also show the high effectiveness of simple defenses against those attacks by proposing new defenses that exploit the unique structure of 3D point clouds. Finally, we attempt to explain the effectiveness of the defenses through the intrinsic structures of both the point clouds and the neural network architectures. Overall, we find that networks that process 3D point cloud data are weak to adversarial attacks, but they are also more easily defensible compared to 2D image classifiers. Our investigation will provide the groundwork for future studies on improving the robustness of deep neural networks that handle 3D data.

cs.CV↗

arXiv2019

Adversarial Defense by Stratified Convolutional Sparse Coding

We propose an adversarial defense method that achieves state-of-the-art performance among attack-agnostic adversarial defense methods while also maintaining robustness to input resolution, scale of adversarial perturbation, and scale of dataset size. Based on convolutional sparse coding, we construct a stratified low-dimensional quasi-natural image space that faithfully approximates the natural image space while also removing adversarial perturbations. We introduce a novel Sparse Transformation Layer (STL) in between the input image and the first layer of the neural network to efficiently project images into our quasi-natural image space. Our experiments show state-of-the-art performance of our method compared to other attack-agnostic adversarial defense methods in various adversarial settings.

cs.CV↗

arXiv2019

Robust magnetoresistance in TaAs2 under pressure up to about 37 GPa

The extremely large magnetoresistance (XMR) in nonmagnetic semimetals has inspired growing interest owing to both intriguing physics and potential applications. We report results of synchrotron X-ray diffraction (SXRD) and electrical transport measurements on TaAs2 under pressure up to ~ 37 GPa, which revealed an anisotropic compression of the unit cell, formation of unusual As-As bonds above 9.5 GPa, and enhancement of metallicity. Interestingly, the MR of TaAs2 under pressure changed gently, which at 1.7 GPa is 96.6% and at 36.6 GPa is still 36.7%. The almost robust MR under pressure could be related to the nearly stable electronic structure unveiled by the ab initio calculations. The discovery would expand the potential use of XMR even under high pressure.

cond-mat.mtrl-sci↗

arXiv2019

Quantum oscillations and nontrivial topological state in a compensated semimetal TaP2

We report systematic magneto-transport measurements and ab initio calculations on single-crystalline TaP2, a new member of the transition-metal dipnictides. We observed unsaturated magnetoresistance (MR) reaching ~ 700% at a magnetic field (B) of 9 T at 2 K along with striking Shubnikov-de Hass (SdH) oscillations. Our analysis on the SdH oscillations reveals nonzero Berry phase, indicating nontrivial band topology. The analysis also uncovers three fundamental magnetic oscillation frequencies of 72 T, 237 T, and 356 T, consistent with the theoretical calculations which reveal one hole pocket and two electron pockets at the L point and one electron pocket at the Z point of the Brillouin zone. We also found negative longitudinal MR (n-MR) within a narrow window of the angles between B and the electric current (I). The n-MR could be fitted with the Adler-Bell-Jackiw chiral anomaly equation but the origin remains yet ambiguous. The ab inito calculations suggest TaP2 as a weak topological insulator with the Z2 indices of (0; 111), which exhibits topological surface states on the (001) surface.

cond-mat.str-el↗

arXiv2019

Magnetic exchange induced Weyl state in a semimetal EuCd2Sb2

Magnetic Weyl semimetals (WSMs) bearing long-time pursuing are still very rare. We herein identified magnetic exchange induced Weyl state in EuCd2Sb2, a semimetal in type IV magnetic space group, via performing high magnetic field (B) magneto-transport measurements and ab initio calculations. For the A-type antiferromagnetic (AFM) structure of EuCd2Sb2, external B larger than 3.2 T can align Eu spins to be fully polarized along the c-axis and consequently drive the system into a ferromagnetic (FM) state. Measurements up to B ~ 55 T revealed a striking Shubnikov-de Hass oscillation imposed by a nontrivial Berry phase. We unveiled a phase transition from a small-gap AFM topological insulator into a FM WSM in which Weyl points emerged along the Γ-Z path. Fermi arcs on (100) and (010) surfaces are also revealed. The results pave a way towards realization of various topological states in a single material through magnetic exchange manipulation.

cond-mat.str-el↗

arXiv2019

How to pick the domain randomization parameters for sim-to-real transfer of reinforcement learning policies?

Recently, reinforcement learning (RL) algorithms have demonstrated remarkable success in learning complicated behaviors from minimally processed input. However, most of this success is limited to simulation. While there are promising successes in applying RL algorithms directly on real systems, their performance on more complex systems remains bottle-necked by the relative data inefficiency of RL algorithms. Domain randomization is a promising direction of research that has demonstrated impressive results using RL algorithms to control real robots. At a high level, domain randomization works by training a policy on a distribution of environmental conditions in simulation. If the environments are diverse enough, then the policy trained on this distribution will plausibly generalize to the real world. A human-specified design choice in domain randomization is the form and parameters of the distribution of simulated environments. It is unclear how to the best pick the form and parameters of this distribution and prior work uses hand-tuned distributions. This extended abstract demonstrates that the choice of the distribution plays a major role in the performance of the trained policies in the real world and that the parameter of this distribution can be optimized to maximize the performance of the trained policies in the real world

cs.LG↗

arXiv2019

A Soft High Force Hand Exoskeleton for Rehabilitation and Assistance of Spinal Cord Injury and Stroke Individuals

Individuals with spinal cord injury (SCI) and stroke who is lack of manipulation capability have a particular need for robotic hand exoskeletons. Among assistive and rehabilitative medical exoskeletons, there exists a sharp trade-off between device power on the one hand and ergonomics and portability on other, devices that provide stronger grasping assistance do so at the cost of patient comfort. This paper proposes using fin-ray inspired, cable-driven finger orthoses to generate high fingertip forces without the painful compressive and shear stresses commonly associated with conventional cable-drive exoskeletons. With combination cable-driven transmission and segmented-finger orthoses, the exoskeleton transmitted larger forces and applied torques discretely to the fingers, leading to strong fingertip forces. A prototype of the finger orthoses and associated cable transmission was fabricated, and force transmission tests of the prototype in the finger flexion mode demonstrated a 2:1 input-output ratio between cable tension and fingertip force, with a maximum fingertip force of 22 N. Moreover, the proposed design provides a comfortable experience for wearers thanks to its lightweight and conformal properties to the hands.

cs.RO↗

arXiv2019

Comfort-Centered Design of a Lightweight and Backdrivable Knee Exoskeleton

This paper presents design principles for comfort-centered wearable robots and their application in a lightweight and backdrivable knee exoskeleton. The mitigation of discomfort is treated as mechanical design and control issues and three solutions are proposed in this paper: 1) a new wearable structure optimizes the strap attachment configuration and suit layout to ameliorate excessive shear forces of conventional wearable structure design; 2) rolling knee joint and double-hinge mechanisms reduce the misalignment in the sagittal and frontal plane, without increasing the mechanical complexity and inertia, respectively; 3) a low impedance mechanical transmission reduces the reflected inertia and damping of the actuator to human, thus the exoskeleton is highly-backdrivable. Kinematic simulations demonstrate that misalignment between the robot joint and knee joint can be reduced by 74% at maximum knee flexion. In experiments, the exoskeleton in the unpowered mode exhibits 1.03 Nm root mean square (RMS) low resistive torque. The torque control experiments demonstrate 0.31 Nm RMS torque tracking error in three human subjects.

cs.RO↗

arXiv2019

DeepSpline: Data-Driven Reconstruction of Parametric Curves and Surfaces

Reconstruction of geometry based on different input modes, such as images or point clouds, has been instrumental in the development of computer aided design and computer graphics. Optimal implementations of these applications have traditionally involved the use of spline-based representations at their core. Most such methods attempt to solve optimization problems that minimize an output-target mismatch. However, these optimization techniques require an initialization that is close enough, as they are local methods by nature. We propose a deep learning architecture that adapts to perform spline fitting tasks accordingly, providing complementary results to the aforementioned traditional methods. We showcase the performance of our approach, by reconstructing spline curves and surfaces based on input images or point clouds.

cs.CV↗

arXiv2018

A Main/Subsidiary Network Framework for Simplifying Binary Neural Network

To reduce memory footprint and run-time latency, techniques such as neural network pruning and binarization have been explored separately. However, it is unclear how to combine the best of the two worlds to get extremely small and efficient models. In this paper, we, for the first time, define the filter-level pruning problem for binary neural networks, which cannot be solved by simply migrating existing structural pruning methods for full-precision models. A novel learning-based approach is proposed to prune filters in our main/subsidiary network framework, where the main network is responsible for learning representative features to optimize the prediction performance, and the subsidiary component works as a filter selector on the main network. To avoid gradient mismatch when training the subsidiary component, we propose a layer-wise and bottom-up scheme. We also provide the theoretical and experimental comparison between our learning-based and greedy rule-based methods. Finally, we empirically demonstrate the effectiveness of our approach applied on several binary models, including binarized NIN, VGG-11, and ResNet-18, on various image classification datasets.

cs.CV↗

arXiv2018

PartNet: A Large-scale Benchmark for Fine-grained and Hierarchical Part-level 3D Object Understanding

We present PartNet: a consistent, large-scale dataset of 3D objects annotated with fine-grained, instance-level, and hierarchical 3D part information. Our dataset consists of 573,585 part instances over 26,671 3D models covering 24 object categories. This dataset enables and serves as a catalyst for many tasks such as shape analysis, dynamic 3D scene modeling and simulation, affordance analysis, and others. Using our dataset, we establish three benchmarking tasks for evaluating 3D part recognition: fine-grained semantic segmentation, hierarchical semantic segmentation, and instance segmentation. We benchmark four state-of-the-art 3D deep learning algorithms for fine-grained semantic segmentation and three baseline methods for hierarchical semantic segmentation. We also propose a novel method for part instance segmentation and demonstrate its superior performance over existing methods.

cs.CV↗

arXiv2018

Binary Ensemble Neural Network: More Bits per Network or More Networks per Bit?

Binary neural networks (BNN) have been studied extensively since they run dramatically faster at lower memory and power consumption than floating-point networks, thanks to the efficiency of bit operations. However, contemporary BNNs whose weights and activations are both single bits suffer from severe accuracy degradation. To understand why, we investigate the representation ability, speed and bias/variance of BNNs through extensive experiments. We conclude that the error of BNNs is predominantly caused by the intrinsic instability (training time) and non-robustness (train & test time). Inspired by this investigation, we propose the Binary Ensemble Neural Network (BENN) which leverages ensemble methods to improve the performance of BNNs with limited efficiency cost. While ensemble techniques have been broadly believed to be only marginally helpful for strong classifiers such as deep neural networks, our analyses and experiments show that they are naturally a perfect fit to boost BNNs. We find that our BENN, which is faster and much more robust than state-of-the-art binary networks, can even surpass the accuracy of the full-precision floating number network with the same architecture.

cs.LG↗