Paper Group ANR 1100
Gradient-Aware Model-based Policy Search. Quantifying the alignment of graph and features in deep learning. Incremental Visual-Inertial 3D Mesh Generation with Structural Regularities. On the Need for Topology-Aware Generative Models for Manifold-Based Defenses. Closing the gap towards end-to-end autonomous vehicle system. Quantified Constraint Han …
Gradient-Aware Model-based Policy Search
| Title | Gradient-Aware Model-based Policy Search |
| Authors | Pierluca D’Oro, Alberto Maria Metelli, Andrea Tirinzoni, Matteo Papini, Marcello Restelli |
| Abstract | Traditional model-based reinforcement learning approaches learn a model of the environment dynamics without explicitly considering how it will be used by the agent. In the presence of misspecified model classes, this can lead to poor estimates, as some relevant available information is ignored. In this paper, we introduce a novel model-based policy search approach that exploits the knowledge of the current agent policy to learn an approximate transition model, focusing on the portions of the environment that are most relevant for policy improvement. We leverage a weighting scheme, derived from the minimization of the error on the model-based policy gradient estimator, in order to define a suitable objective function that is optimized for learning the approximate transition model. Then, we integrate this procedure into a batch policy improvement algorithm, named Gradient-Aware Model-based Policy Search (GAMPS), which iteratively learns a transition model and uses it, together with the collected trajectories, to compute the new policy parameters. Finally, we empirically validate GAMPS on benchmark domains analyzing and discussing its properties. |
| Tasks | |
| Published | 2019-09-09 |
| URL | https://arxiv.org/abs/1909.04115v2 |
| https://arxiv.org/pdf/1909.04115v2.pdf | |
| PWC | https://paperswithcode.com/paper/gradient-aware-model-based-policy-search |
| Repo | |
| Framework | |
Quantifying the alignment of graph and features in deep learning
| Title | Quantifying the alignment of graph and features in deep learning |
| Authors | Yifan Qian, Paul Expert, Tom Rieu, Pietro Panzarasa, Mauricio Barahona |
| Abstract | We show that the classification performance of Graph Convolutional Networks is related to the alignment between features, graph and ground truth, which we quantify using a subspace alignment measure corresponding to the Frobenius norm of the matrix of pairwise chordal distances between three subspaces associated with features, graph and ground truth. The proposed measure is based on the principal angles between subspaces and has both spectral and geometrical interpretations. We showcase the relationship between the subspace alignment measure and the classification performance through the study of limiting cases of Graph Convolutional Networks as well as systematic randomizations of both features and graph structure applied to a constructive example and several examples of citation networks of different origin. The analysis also reveals the relative importance of the graph and features for classification purposes. |
| Tasks | |
| Published | 2019-05-30 |
| URL | https://arxiv.org/abs/1905.12921v1 |
| https://arxiv.org/pdf/1905.12921v1.pdf | |
| PWC | https://paperswithcode.com/paper/quantifying-the-alignment-of-graph-and |
| Repo | |
| Framework | |
Incremental Visual-Inertial 3D Mesh Generation with Structural Regularities
| Title | Incremental Visual-Inertial 3D Mesh Generation with Structural Regularities |
| Authors | Antoni Rosinol, Torsten Sattler, Marc Pollefeys, Luca Carlone |
| Abstract | Visual-Inertial Odometry (VIO) algorithms typically rely on a point cloud representation of the scene that does not model the topology of the environment. A 3D mesh instead offers a richer, yet lightweight, model. Nevertheless, building a 3D mesh out of the sparse and noisy 3D landmarks triangulated by a VIO algorithm often results in a mesh that does not fit the real scene. In order to regularize the mesh, previous approaches decouple state estimation from the 3D mesh regularization step, and either limit the 3D mesh to the current frame or let the mesh grow indefinitely. We propose instead to tightly couple mesh regularization and state estimation by detecting and enforcing structural regularities in a novel factor-graph formulation. We also propose to incrementally build the mesh by restricting its extent to the time-horizon of the VIO optimization; the resulting 3D mesh covers a larger portion of the scene than a per-frame approach while its memory usage and computational complexity remain bounded. We show that our approach successfully regularizes the mesh, while improving localization accuracy, when structural regularities are present, and remains operational in scenes without regularities. |
| Tasks | |
| Published | 2019-03-04 |
| URL | https://arxiv.org/abs/1903.01067v2 |
| https://arxiv.org/pdf/1903.01067v2.pdf | |
| PWC | https://paperswithcode.com/paper/incremental-visual-inertial-3d-mesh |
| Repo | |
| Framework | |
On the Need for Topology-Aware Generative Models for Manifold-Based Defenses
| Title | On the Need for Topology-Aware Generative Models for Manifold-Based Defenses |
| Authors | Uyeong Jang, Susmit Jha, Somesh Jha |
| Abstract | Machine-learning (ML) algorithms or models, especially deep neural networks (DNNs), have shown significant promise in several areas. However, researchers have recently demonstrated that ML algorithms, especially DNNs, are vulnerable to adversarial examples (slightly perturbed samples that cause misclassification). The existence of adversarial examples has hindered the deployment of ML algorithms in safety-critical sectors, such as security. Several defenses for adversarial examples exist in the literature. One of the important classes of defenses are manifold-based defenses, where a sample is ``pulled back” into the data manifold before classifying. These defenses rely on the assumption that data lie in a manifold of a lower dimension than the input space. These defenses use a generative model to approximate the input distribution. In this paper, we investigate the following question: do the generative models used in manifold-based defenses need to be topology-aware? We suggest the answer is yes, and we provide theoretical and empirical evidence to support our claim. | |
| Tasks | Data Augmentation |
| Published | 2019-09-07 |
| URL | https://arxiv.org/abs/1909.03334v4 |
| https://arxiv.org/pdf/1909.03334v4.pdf | |
| PWC | https://paperswithcode.com/paper/on-need-for-topology-awareness-of-generative |
| Repo | |
| Framework | |
Closing the gap towards end-to-end autonomous vehicle system
| Title | Closing the gap towards end-to-end autonomous vehicle system |
| Authors | Yonatan Glassner, Liran Gispan, Ariel Ayash, Tal Furman Shohet |
| Abstract | Designing a driving policy for autonomous vehicles is a difficult task. Recent studies suggested an end-toend (E2E) training of a policy to predict car actuators directly from raw sensory inputs. It is appealing due to the ease of labeled data collection and since handcrafted features are avoided. Explicit drawbacks such as interpretability, safety enforcement and learning efficiency limit the practical application of the approach. In this paper, we amend the basic E2E architecture to address these shortcomings, while retaining the power of end-to-end learning. A key element in our proposed architecture is formulation of the learning problem as learning of trajectory. We also apply a Gaussian mixture model loss to contend with multi-modal data, and adopt a finance risk measure, conditional value at risk, to emphasize rare events. We analyze the effect of each concept and present driving performance in a highway scenario in the TORCS simulator. Video is available in this link: https://www.youtube.com/watch?v=1JYNBZNOe_4 |
| Tasks | Autonomous Vehicles |
| Published | 2019-01-01 |
| URL | http://arxiv.org/abs/1901.00114v2 |
| http://arxiv.org/pdf/1901.00114v2.pdf | |
| PWC | https://paperswithcode.com/paper/closing-the-gap-towards-end-to-end-autonomous |
| Repo | |
| Framework | |
Quantified Constraint Handling Rules
| Title | Quantified Constraint Handling Rules |
| Authors | Vincent Barichard, Igor Stéphan |
| Abstract | We shift the QCSP (Quantified Constraint Satisfaction Problems) framework to the QCHR (Quantified Constraint Handling Rules) framework by enabling dynamic binder and access to user-defined constraints. QCSP offers a natural framework to express PSPACE problems as finite two-players games. But to define a QCSP model, the binder must be formerly known and cannot be built dynamically even if the worst case won’t occur. To overcome this issue, we define the new QCHR formalism that allows to build the binder dynamically during the solving. Our QCHR models exhibit state-of-the-art performances on static binder and outperforms previous QCSP approaches when the binder is dynamic. |
| Tasks | |
| Published | 2019-09-18 |
| URL | https://arxiv.org/abs/1909.08243v1 |
| https://arxiv.org/pdf/1909.08243v1.pdf | |
| PWC | https://paperswithcode.com/paper/quantified-constraint-handling-rules |
| Repo | |
| Framework | |
The Unconstrained Ear Recognition Challenge 2019 - ArXiv Version With Appendix
| Title | The Unconstrained Ear Recognition Challenge 2019 - ArXiv Version With Appendix |
| Authors | Žiga Emeršič, Aruna Kumar S. V., B. S. Harish, Weronika Gutfeter, Jalil Nourmohammadi Khiarak, Andrzej Pacut, Earnest Hansley, Mauricio Pamplona Segundo, Sudeep Sarkar, Hyeonjung Park, Gi Pyo Nam, Ig-Jae Kim, Sagar G. Sangodkar, Ümit Kaçar, Murvet Kirci, Li Yuan, Jishou Yuan, Haonan Zhao, Fei Lu, Junying Mao, Xiaoshuang Zhang, Dogucan Yaman, Fevziye Irem Eyiokur, Kadir Bulut Özler, Hazım Kemal Ekenel, Debbrota Paul Chowdhury, Sambit Bakshi, Pankaj K. Sa, Banshidhar Majhi, Peter Peer, Vitomir Štruc |
| Abstract | This paper presents a summary of the 2019 Unconstrained Ear Recognition Challenge (UERC), the second in a series of group benchmarking efforts centered around the problem of person recognition from ear images captured in uncontrolled settings. The goal of the challenge is to assess the performance of existing ear recognition techniques on a challenging large-scale ear dataset and to analyze performance of the technology from various viewpoints, such as generalization abilities to unseen data characteristics, sensitivity to rotations, occlusions and image resolution and performance bias on sub-groups of subjects, selected based on demographic criteria, i.e. gender and ethnicity. Research groups from 12 institutions entered the competition and submitted a total of 13 recognition approaches ranging from descriptor-based methods to deep-learning models. The majority of submissions focused on ensemble based methods combining either representations from multiple deep models or hand-crafted with learned image descriptors. Our analysis shows that methods incorporating deep learning models clearly outperform techniques relying solely on hand-crafted descriptors, even though both groups of techniques exhibit similar behaviour when it comes to robustness to various covariates, such presence of occlusions, changes in (head) pose, or variability in image resolution. The results of the challenge also show that there has been considerable progress since the first UERC in 2017, but that there is still ample room for further research in this area. |
| Tasks | Person Recognition |
| Published | 2019-03-11 |
| URL | http://arxiv.org/abs/1903.04143v3 |
| http://arxiv.org/pdf/1903.04143v3.pdf | |
| PWC | https://paperswithcode.com/paper/the-unconstrained-ear-recognition-challenge-1 |
| Repo | |
| Framework | |
Data-Efficient Learning for Sim-to-Real Robotic Grasping using Deep Point Cloud Prediction Networks
| Title | Data-Efficient Learning for Sim-to-Real Robotic Grasping using Deep Point Cloud Prediction Networks |
| Authors | Xinchen Yan, Mohi Khansari, Jasmine Hsu, Yuanzheng Gong, Yunfei Bai, Sören Pirk, Honglak Lee |
| Abstract | Training a deep network policy for robot manipulation is notoriously costly and time consuming as it depends on collecting a significant amount of real world data. To work well in the real world, the policy needs to see many instances of the task, including various object arrangements in the scene as well as variations in object geometry, texture, material, and environmental illumination. In this paper, we propose a method that learns to perform table-top instance grasping of a wide variety of objects while using no real world grasping data, outperforming the baseline using 2.5D shape by 10%. Our method learns 3D point cloud of object, and use that to train a domain-invariant grasping policy. We formulate the learning process as a two-step procedure: 1) Learning a domain-invariant 3D shape representation of objects from about 76K episodes in simulation and about 530 episodes in the real world, where each episode lasts less than a minute and 2) Learning a critic grasping policy in simulation only based on the 3D shape representation from step 1. Our real world data collection in step 1 is both cheaper and faster compared to existing approaches as it only requires taking multiple snapshots of the scene using a RGBD camera. Finally, the learned 3D representation is not specific to grasping, and can potentially be used in other interaction tasks. |
| Tasks | 3D Shape Representation, Robotic Grasping |
| Published | 2019-06-21 |
| URL | https://arxiv.org/abs/1906.08989v1 |
| https://arxiv.org/pdf/1906.08989v1.pdf | |
| PWC | https://paperswithcode.com/paper/data-efficient-learning-for-sim-to-real |
| Repo | |
| Framework | |
Studying the Impact of Mood on Identifying Smartphone Users
| Title | Studying the Impact of Mood on Identifying Smartphone Users |
| Authors | Khadija Zanna, Sayde King, Tempestt Neal, Shaun Canavan |
| Abstract | This paper explores the identification of smartphone users when certain samples collected while the subject felt happy, upset or stressed were absent or present. We employ data from 19 subjects using the StudentLife dataset, a dataset collected by researchers at Dartmouth College that was originally collected to correlate behaviors characterized by smartphone usage patterns with changes in stress and academic performance. Although many previous works on behavioral biometrics have implied that mood is a source of intra-person variation which may impact biometric performance, our results contradict this assumption. Our findings show that performance worsens when removing samples that were generated when subjects may be happy, upset, or stressed. Thus, there is no indication that mood negatively impacts performance. However, we do find that changes existing in smartphone usage patterns may correlate with mood, including changes in locking, audio, location, calling, homescreen, and e-mail habits. Thus, we show that while mood is a source of intra-person variation, it may be an inaccurate assumption that biometric systems (particularly, mobile biometrics) are likely influenced by mood. |
| Tasks | |
| Published | 2019-06-27 |
| URL | https://arxiv.org/abs/1906.11960v1 |
| https://arxiv.org/pdf/1906.11960v1.pdf | |
| PWC | https://paperswithcode.com/paper/studying-the-impact-of-mood-on-identifying |
| Repo | |
| Framework | |
Partial-Order, Partially-Seen Observations of Fluents or Actions for Plan Recognition as Planning
| Title | Partial-Order, Partially-Seen Observations of Fluents or Actions for Plan Recognition as Planning |
| Authors | Jennifer M. Nelson, Rogelio E. Cardona-Rivera |
| Abstract | This work aims to make plan recognition as planning more ready for real-world scenarios by adapting previous compilations to work with partial-order, half-seen observations of both fluents and actions. We first redefine what observations can be and what it means to satisfy each kind. We then provide a compilation from plan recognition problem to classical planning problem, similar to original work by Ramirez and Geffner, but accommodating these more complex observation types. This compilation can be adapted towards other planning-based plan recognition techniques. Lastly we evaluate this method against an “ignore complexity” strategy that uses the original method by Ramirez and Geffner. Our experimental results suggest that, while slower, our method is equally or more accurate than baseline methods; our technique sometimes significantly reduces the size of the solution to the plan recognition problem, i.e, the size of the optimal goal set. We discuss these findings in the context of plan recognition problem difficulty and present an avenue for future work. |
| Tasks | |
| Published | 2019-11-14 |
| URL | https://arxiv.org/abs/1911.05876v1 |
| https://arxiv.org/pdf/1911.05876v1.pdf | |
| PWC | https://paperswithcode.com/paper/partial-order-partially-seen-observations-of |
| Repo | |
| Framework | |
Feature Selection for multi-labeled variables via Dependency Maximization
| Title | Feature Selection for multi-labeled variables via Dependency Maximization |
| Authors | Salimeh Yasaei Sekeh, Alfred O. Hero |
| Abstract | Feature selection and reducing the dimensionality of data is an essential step in data analysis. In this work, we propose a new criterion for feature selection that is formulated as conditional information between features given the labeled variable. Instead of using the standard mutual information measure based on Kullback-Leibler divergence, we use our proposed criterion to filter out redundant features for the purpose of multiclass classification. This approach results in an efficient and fast non-parametric implementation of feature selection as it can be directly estimated using a geometric measure of dependency, the global Friedman-Rafsky (FR) multivariate run test statistic constructed by a global minimal spanning tree (MST). We demonstrate the advantages of our proposed feature selection approach through simulation. In addition the proposed feature selection method is applied to the MNIST data set. |
| Tasks | Feature Selection |
| Published | 2019-02-10 |
| URL | https://arxiv.org/abs/1902.03544v3 |
| https://arxiv.org/pdf/1902.03544v3.pdf | |
| PWC | https://paperswithcode.com/paper/feature-selection-for-multi-labeled-variables |
| Repo | |
| Framework | |
Automated Definition of Skeletal Disease Burden in Metastatic Prostate Carcinoma: a 3D analysis of SPECT/CT images
| Title | Automated Definition of Skeletal Disease Burden in Metastatic Prostate Carcinoma: a 3D analysis of SPECT/CT images |
| Authors | Francesco Fiz, Helmut Dittmann, Cristina Campi, Matthias Weissinger, Samine Sahbai, Matthias Reimold, Arnulf Stenzl, Michele Piana, Gianmario Sambuceti, Christian la Fougère |
| Abstract | To meet the current need for skeletal tumor-load estimation in prostate cancer (mCRPC), we developed a novel approach, based on adaptive bone segmentation. In this study, we compared the program output with existing estimates and with the radiological outcome. Seventy-six whole-body 99mTc-DPD-SPECT/CT from mCRPC patients were analyzed. The software identified the whole skeletal volume (SVol) and classified it voxels metastases (MVol) or normal bone (BVol). SVol was compared with the estimation of a commercial software. MVol was compared with manual assessment and with PSA-level. Counts/voxel were extracted from MVol and BVol. After six cycles of 223RaCl2-therapy every patient was re-evaluated as progressing (PD), stabilized (SD) or responsive (PR). SVol correlated with the one of the commercial software (R=0,99, p<0,001). MVol correlated with manually-counted lesions (R=0,61, p<0,001) and PSA (R=0,46, p<0.01). PD had a lower counts/voxel in MVol than PR/SD (715 \pm 190 Vs. 975 \pm 215 and 1058 \pm 255, p<0,05 and p<0,01) and in BVol (PD 275 \pm 60, PR 515 \pm 188 and SD 528 \pm 162 counts/voxel, p<0,001). Segmentation-based tumor load correlated with radiological/laboratory indices. Uptake was linked with the clinical outcome, suggesting that metastases in PD patients have a lower affinity for bone-seeking radionuclides and might benefit less from bone-targeted radioisotope therapies. |
| Tasks | |
| Published | 2019-06-19 |
| URL | https://arxiv.org/abs/1906.08200v1 |
| https://arxiv.org/pdf/1906.08200v1.pdf | |
| PWC | https://paperswithcode.com/paper/automated-definition-of-skeletal-disease |
| Repo | |
| Framework | |
Prune Sampling: a MCMC inference technique for discrete and deterministic Bayesian networks
| Title | Prune Sampling: a MCMC inference technique for discrete and deterministic Bayesian networks |
| Authors | Frank Phillipson, Jurriaan Parie, Ron Weikamp |
| Abstract | We introduce and characterise the performance of the Markov chain Monte Carlo (MCMC) inference method Prune Sampling for discrete and deterministic Bayesian networks (BNs). We developed a procedure to obtain the performance of a MCMC sampling method in the limit of infinite simulation time, extrapolated from relatively short simulations. This approach was used to conduct a study to compare the accuracy, rate of convergence and the time consumption of Prune Sampling with two conventional MCMC sampling methods: Gibbs- and Metropolis sampling. We show that Markov chains created by Prune Sampling always converge to the desired posterior distribution, also for networks where conventional Gibbs sampling fails. Beside this, we demonstrate that pruning outperforms Gibbs sampling, at least for a certain class of BNs. Though, this tempting feature comes at a price. In the first version of Prune Sampling, for large BNs the procedure to choose the next iteration step uniformly is rather time intensive. Our conclusion is that Prune Sampling is a competitive method for all types of small and medium sized BNs, but (for now) standard methods still perform better for all types of large BNs. |
| Tasks | |
| Published | 2019-08-17 |
| URL | https://arxiv.org/abs/1908.06335v1 |
| https://arxiv.org/pdf/1908.06335v1.pdf | |
| PWC | https://paperswithcode.com/paper/prune-sampling-a-mcmc-inference-technique-for |
| Repo | |
| Framework | |
Robot Capability and Intention in Trust-based Decisions across Tasks
| Title | Robot Capability and Intention in Trust-based Decisions across Tasks |
| Authors | Yaqi Xie, Indu P Bodala, Desmond C. Ong, David Hsu, Harold Soh |
| Abstract | In this paper, we present results from a human-subject study designed to explore two facets of human mental models of robots—inferred capability and intention—and their relationship to overall trust and eventual decisions. In particular, we examine delegation situations characterized by uncertainty, and explore how inferred capability and intention are applied across different tasks. We develop an online survey where human participants decide whether to delegate control to a simulated UAV agent. Our study shows that human estimations of robot capability and intent correlate strongly with overall self-reported trust. However, overall trust is not independently sufficient to determine whether a human will decide to trust (delegate) a given task to a robot. Instead, our study reveals that estimations of robot intention, capability, and overall trust are integrated when deciding to delegate. From a broader perspective, these results suggest that calibrating overall trust alone is insufficient; to make correct decisions, humans need (and use) multi-faceted mental models when collaborating with robots across multiple contexts. |
| Tasks | |
| Published | 2019-09-03 |
| URL | https://arxiv.org/abs/1909.05329v1 |
| https://arxiv.org/pdf/1909.05329v1.pdf | |
| PWC | https://paperswithcode.com/paper/robot-capability-and-intention-in-trust-based |
| Repo | |
| Framework | |
Interplanetary Transfers via Deep Representations of the Optimal Policy and/or of the Value Function
| Title | Interplanetary Transfers via Deep Representations of the Optimal Policy and/or of the Value Function |
| Authors | Dario Izzo, Ekin Öztürk, Marcus Märtens |
| Abstract | A number of applications to interplanetary trajectories have been recently proposed based on deep networks. These approaches often rely on the availability of a large number of optimal trajectories to learn from. In this paper we introduce a new method to quickly create millions of optimal spacecraft trajectories from a single nominal trajectory. Apart from the generation of the nominal trajectory, no additional optimal control problems need to be solved as all the trajectories, by construction, satisfy Pontryagin’s minimum principle and the relevant transversality conditions. We then consider deep feed forward neural networks and benchmark three learning methods on the created dataset: policy imitation, value function learning and value function gradient learning. Our results are shown for the case of the interplanetary trajectory optimization problem of reaching Venus orbit, with the nominal trajectory starting from the Earth. We find that both policy imitation and value function gradient learning are able to learn the optimal state feedback, while in the case of value function learning the optimal policy is not captured, only the final value of the optimal propellant mass is. |
| Tasks | |
| Published | 2019-04-18 |
| URL | http://arxiv.org/abs/1904.08809v1 |
| http://arxiv.org/pdf/1904.08809v1.pdf | |
| PWC | https://paperswithcode.com/paper/interplanetary-transfers-via-deep |
| Repo | |
| Framework | |
