We introduce a practical pipeline that interactively encodes multimodal human demonstrations for robot teaching. This pipeline is designed as an input system for a framework called Learning-from-Observation (LfO), which aims to program household robots with manipulative tasks through few-shots human demonstration without coding. While most previous LfO systems run with visual demonstration, recent research on robot teaching has shown the effectiveness of verbal instruction in making recognition robust and teaching interactive. To the best of our knowledge, however, no LfO system has yet been proposed that utilizes both verbal instruction and interaction, namely \textit{multimodal LfO}. This paper proposes the interactive task encoding system (ITES) as an input pipeline for multimodal LfO. ITES assumes that the user teaches step-by-step, pausing hand movements in order to match the granularity of human instructions with the granularity of robot execution. ITES recognizes tasks based on step-by-step verbal instructions that accompany the hand movements. Additionally, the recognition is made robust through interactions with the user. We test ITES on a real robot and show that the user can successfully teach multiple operations through multimodal demonstrations. The results suggest the usefulness of ITES for multimodal LfO. The source code is available at //github.com/microsoft/symbolic-robot-teaching-interface.
相關內容
Automatic Robotic Assembly Sequence Planning (RASP) can significantly improve productivity and resilience in modern manufacturing along with the growing need for greater product customization. One of the main challenges in realizing such automation resides in efficiently finding solutions from a growing number of potential sequences for increasingly complex assemblies. Besides, costly feasibility checks are always required for the robotic system. To address this, we propose a holistic graphical approach including a graph representation called Assembly Graph for product assemblies and a policy architecture, Graph Assembly Processing Network, dubbed GRACE for assembly sequence generation. Secondly, we use GRACE to extract meaningful information from the graph input and predict assembly sequences in a step-by-step manner. In experiments, we show that our approach can predict feasible assembly sequences across product variants of aluminum profiles based on data collected in simulation of a dual-armed robotic system. We further demonstrate that our method is capable of detecting infeasible assemblies, substantially alleviating the undesirable impacts from false predictions, and hence facilitating real-world deployment soon. Code and training data will be open-sourced.
The focus of this work is to investigate unsupervised approaches to overcome quintessential challenges in designing task-oriented dialog schema: assigning intent labels to each dialog turn (intent clustering) and generating a set of intents based on the intent clustering methods (intent induction). We postulate there are two salient factors for automatic induction of intents: (1) clustering algorithm for intent labeling and (2) user utterance embedding space. We compare existing off-the-shelf clustering models and embeddings based on DSTC11 evaluation. Our extensive experiments demonstrate that the combined selection of utterance embedding and clustering method in the intent induction task should be carefully considered. We also present that pretrained MiniLM with Agglomerative clustering shows significant improvement in NMI, ARI, F1, accuracy and example coverage in intent induction tasks. The source codes are available at //github.com/Jeiyoon/dstc11-track2.
3D scene understanding plays a vital role in vision-based autonomous driving. While most existing methods focus on 3D object detection, they have difficulty describing real-world objects of arbitrary shapes and infinite classes. Towards a more comprehensive perception of a 3D scene, in this paper, we propose a SurroundOcc method to predict the 3D occupancy with multi-camera images. We first extract multi-scale features for each image and adopt spatial 2D-3D attention to lift them to the 3D volume space. Then we apply 3D convolutions to progressively upsample the volume features and impose supervision on multiple levels. To obtain dense occupancy prediction, we design a pipeline to generate dense occupancy ground truth without expansive occupancy annotations. Specifically, we fuse multi-frame LiDAR scans of dynamic objects and static scenes separately. Then we adopt Poisson Reconstruction to fill the holes and voxelize the mesh to get dense occupancy labels. Extensive experiments on nuScenes and SemanticKITTI datasets demonstrate the superiority of our method. Code and dataset are available at //github.com/weiyithu/SurroundOcc
The complexity of today's robot control systems implies difficulty in developing them efficiently and reliably. Systems engineering (SE) and frameworks come to help. The framework metamodels are needed to support the standardisation and correctness of the created application models. Although the use of frameworks is widespread nowadays, for the most popular of them, Robot Operating System (ROS) version 1, a complete, contemporary metamodel, has been missing so far. This article proposes a new metamodel for ROS (MeROS), which addresses both the running system and developer workspace. For compatibility with the latest versions of ROS 1, the metamodel includes the latest ROS 1 concepts such as nodelet, action, and metapackage. An essential addition to the original ROS concepts is the grouping concepts, which provide an opportunity to illustrate the decomposition of the system, as well as varying degrees of detail in its presentation. The metamodel is derived from the requirements and then verified on the practical example of the Rico assistive robot. The matter is described in the SysML language, supported by standard development tools to conduct projects in the spirit of SE.
Genito-Pelvic Pain/Penetration-Disorder (GPPPD) is a common disorder but rarely treated in routine care. Previous research documents that GPPPD symptoms can be treated effectively using internet-based psychological interventions. However, non-response remains common for all state-of-the-art treatments and it is unclear which patient groups are expected to benefit most from an internet-based intervention. Multivariable prediction models are increasingly used to identify predictors of heterogeneous treatment effects, and to allocate treatments with the greatest expected benefits. In this study, we developed and internally validated a multivariable decision tree model that predicts effects of an internet-based treatment on a multidimensional composite score of GPPPD symptoms. Data of a randomized controlled trial comparing the internet-based intervention to a waitlist control group (N =200) was used to develop a decision tree model using model-based recursive partitioning. Model performance was assessed by examining the apparent and bootstrap bias-corrected performance. The final pruned decision tree consisted of one splitting variable, joint dyadic coping, based on which two response clusters emerged. No effect was found for patients with low dyadic coping ($n$=33; $d$=0.12; 95% CI: -0.57-0.80), while large effects ($d$=1.00; 95%CI: 0.68-1.32; $n$=167) are predicted for those with high dyadic coping at baseline. The bootstrap-bias-corrected performance of the model was $R^2$=27.74% (RMSE=13.22).
Image-based table recognition is a challenging task due to the diversity of table styles and the complexity of table structures. Most of the previous methods focus on a non-end-to-end approach which divides the problem into two separate sub-problems: table structure recognition; and cell-content recognition and then attempts to solve each sub-problem independently using two separate systems. In this paper, we propose an end-to-end multi-task learning model for image-based table recognition. The proposed model consists of one shared encoder, one shared decoder, and three separate decoders which are used for learning three sub-tasks of table recognition: table structure recognition, cell detection, and cell-content recognition. The whole system can be easily trained and inferred in an end-to-end approach. In the experiments, we evaluate the performance of the proposed model on two large-scale datasets: FinTabNet and PubTabNet. The experiment results show that the proposed model outperforms the state-of-the-art methods in all benchmark datasets.
Despite the recent progress in deep learning, most approaches still go for a silo-like solution, focusing on learning each task in isolation: training a separate neural network for each individual task. Many real-world problems, however, call for a multi-modal approach and, therefore, for multi-tasking models. Multi-task learning (MTL) aims to leverage useful information across tasks to improve the generalization capability of a model. This thesis is concerned with multi-task learning in the context of computer vision. First, we review existing approaches for MTL. Next, we propose several methods that tackle important aspects of multi-task learning. The proposed methods are evaluated on various benchmarks. The results show several advances in the state-of-the-art of multi-task learning. Finally, we discuss several possibilities for future work.
Catastrophic forgetting refers to the tendency that a neural network "forgets" the previous learned knowledge upon learning new tasks. Prior methods have been focused on overcoming this problem on convolutional neural networks (CNNs), where the input samples like images lie in a grid domain, but have largely overlooked graph neural networks (GNNs) that handle non-grid data. In this paper, we propose a novel scheme dedicated to overcoming catastrophic forgetting problem and hence strengthen continual learning in GNNs. At the heart of our approach is a generic module, termed as topology-aware weight preserving~(TWP), applicable to arbitrary form of GNNs in a plug-and-play fashion. Unlike the main stream of CNN-based continual learning methods that rely on solely slowing down the updates of parameters important to the downstream task, TWP explicitly explores the local structures of the input graph, and attempts to stabilize the parameters playing pivotal roles in the topological aggregation. We evaluate TWP on different GNN backbones over several datasets, and demonstrate that it yields performances superior to the state of the art. Code is publicly available at \url{//github.com/hhliu79/TWP}.
To address the sparsity and cold start problem of collaborative filtering, researchers usually make use of side information, such as social networks or item attributes, to improve recommendation performance. This paper considers the knowledge graph as the source of side information. To address the limitations of existing embedding-based and path-based methods for knowledge-graph-aware recommendation, we propose Ripple Network, an end-to-end framework that naturally incorporates the knowledge graph into recommender systems. Similar to actual ripples propagating on the surface of water, Ripple Network stimulates the propagation of user preferences over the set of knowledge entities by automatically and iteratively extending a user's potential interests along links in the knowledge graph. The multiple "ripples" activated by a user's historically clicked items are thus superposed to form the preference distribution of the user with respect to a candidate item, which could be used for predicting the final clicking probability. Through extensive experiments on real-world datasets, we demonstrate that Ripple Network achieves substantial gains in a variety of scenarios, including movie, book and news recommendation, over several state-of-the-art baselines.
Most previous event extraction studies have relied heavily on features derived from annotated event mentions, thus cannot be applied to new event types without annotation effort. In this work, we take a fresh look at event extraction and model it as a grounding problem. We design a transferable neural architecture, mapping event mentions and types jointly into a shared semantic space using structural and compositional neural networks, where the type of each event mention can be determined by the closest of all candidate types . By leveraging (1)~available manual annotations for a small set of existing event types and (2)~existing event ontologies, our framework applies to new event types without requiring additional annotation. Experiments on both existing event types (e.g., ACE, ERE) and new event types (e.g., FrameNet) demonstrate the effectiveness of our approach. \textit{Without any manual annotations} for 23 new event types, our zero-shot framework achieved performance comparable to a state-of-the-art supervised model which is trained from the annotations of 500 event mentions.
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191