arXiv subjects

Weiwei Gu

Publications and source records attributed to Weiwei Gu.

16 recordsLinked to original sources

arXiv2026

Monkey See, Can Monkey Do? A Benchmark for Evaluating Robot Skill Learning by Observation

Learning from Observation (LfO) is a fundamental robotic capability that replicates how humans and animals socially learn from each other. Beyond its biological parallels, this modality provides a practical solution for data scaling in sample-inefficient and data-starved domains like robotics. Recent work has demonstrated promising results in learning manipulation skills from human videos, yet progress in this area remains difficult to assess. Existing methods vary widely in assumptions, hardware choices, and environment setups making it difficult to draw meaningful comparisons and identify advances in the field. To address these challenges, we introduce RoboReel: a unified benchmark for evaluating models that learn policies from human videos. RoboReel consists of bundled real-world human demonstration videos, simulated robot trajectories, and evaluation environments on ten manipulation tasks. We develop four test suites to evaluate the models' performance on multiple axes, including the robustness to visual distractors and the ability to complete long-horizon tasks. Our benchmark covers learning-from-observation models from different categories, and studies the effectiveness of multiple representation choices in our benchmark evaluation that covers over seven state-of-the-art algorithms (including our VLA based variants) in the field of LfO. Finally, we present an analysis of the different types of algorithms showing that long-horizon tasks and tasks with low tolerances are still challenging for current models. Webpage: https://roboreel.github.io

cs.RO↗

arXiv2026

PokeNet: Learning Kinematic Models of Articulated Objects from Human Observations

Articulation modeling enables robots to learn joint parameters of articulated objects for effective manipulation which can then be used downstream for skill learning or planning. Existing approaches often rely on prior knowledge about the objects, such as the number or type of joints. Some of these approaches also fail to recover occluded joints that are only revealed during interaction. Others require large numbers of multi-view images for every object, which is impractical in real-world settings. Furthermore, prior works neglect the order of manipulations, which is essential for many multi-DoF objects where one joint must be operated before another, such as a dishwasher. We introduce PokeNet, an end-to-end framework that estimates articulation models from a single human demonstration without prior object knowledge. Given a sequence of point cloud observations of a human manipulating an unknown object, PokeNet predicts joint parameters, infers manipulation order, and tracks joint states over time. PokeNet outperforms existing state-of-the-art methods, improving joint axis and state estimation accuracy by an average of over 27% across diverse objects, including novel and unseen categories. We demonstrate these gains in both simulation and real-world environments.

cs.RO↗

arXiv2025

Assessing the Robustness and Reducibility of Multiplex Networks with Embedding-Aided Interlayer Similarities

The study of interlayer similarity of multiplex networks helps to understand the intrinsic structure of complex systems, revealing how changes in one layer can propagate and affect others, thus enabling broad implications for transportation, social, and biological systems. Existing algorithms that measure similarity between network layers typically encode only partial information, which limits their effectiveness in capturing the full complexity inherent in multiplex networks. To address this limitation, we propose a novel interlayer similarity measuring approach named Embedding Aided inTerlayer Similarity (EATSim). EATSim concurrently incorporates intralayer structural similarity and cross-layer anchor node alignment consistency, providing a more comprehensive framework for analyzing interconnected systems. Extensive experiments on both synthetic and real-world networks demonstrate that EATSim effectively captures the underlying geometric similarities between interconnected networks, significantly improving the accuracy of interlayer similarity measurement. Moreover, EATSim achieves state-of-the-art performance in two downstream applications: predicting network robustness and network reducibility, showing its great potential in enhancing the understanding and management of complex systems.

cs.SI↗

arXiv2025

Learning Sequential Kinematic Models from Demonstrations for Multi-Jointed Articulated Objects

As robots become more generalized and deployed in diverse environments, they must interact with complex objects, many with multiple independent joints or degrees of freedom (DoF) requiring precise control. A common strategy is object modeling, where compact state-space models are learned from real-world observations and paired with classical planning. However, existing methods often rely on prior knowledge or focus on single-DoF objects, limiting their applicability. They also fail to handle occluded joints and ignore the manipulation sequences needed to access them. We address this by learning object models from human demonstrations. We introduce Object Kinematic Sequence Machines (OKSMs), a novel representation capturing both kinematic constraints and manipulation order for multi-DoF objects. To estimate these models from point cloud data, we present Pokenet, a deep neural network trained on human demonstrations. We validate our approach on 8,000 simulated and 1,600 real-world annotated samples. Pokenet improves joint axis and state estimation by over 20 percent on real-world data compared to prior methods. Finally, we demonstrate OKSMs on a Sawyer robot using inverse kinematics-based planning to manipulate multi-DoF objects.

cs.RO↗

arXiv2024

Continual Robot Skill and Task Learning via Dialogue

Interactive robot learning is a challenging problem as the robot is present with human users who expect the robot to learn novel skills to solve novel tasks perpetually with sample efficiency. In this work we present a framework for robots to continually learn tasks and visuo-motor skills and query for novel skills via dialog interactions with human users. Our robot agent maintains a skill library, and uses an existing LLM to perform grounded dialog interactions to query unknown skills from real human users. We developed a novel visual-motor control policy Action Chunking Transformer with Low Rank Adaptation (ACT-LoRA) that can continually learn novel skills using only a few demonstrations which is critical in human-robot interaction scenarios. The paper has twin goals: Firstly to demonstrate better continual learning in simulation; and secondly, to demonstrate the use of our dialog based learning framework in a realistic human-robot interaction use case. Our ACT-LoRA policy consistently outperforms a GMM-LoRA baseline on multiple continual learning simulation benchmarks by achieving > 300% improvements on novel skills, while achieving comparable performance in existing skills. Moreover, with our IRB approved human-subjects study we demonstrate that our dialog based continual learning framework allows users to teach robots cooking skills successfully (100%) while spending a higher ratio of time on finishing an auxiliary distraction tasks in the test phase of the study compared to a non-learning language based agent (p < 0.001).

cs.RO↗

arXiv2024

Pay Attention to Weak Ties: A Heterogeneous Multiplex Representation Learning Framework for Link Prediction

Graph neural networks (GNNs) can learn effective node representations that significantly improve link prediction accuracy. However, most GNN-based link prediction algorithms are incompetent to predict weak ties connecting different communities. Most link prediction algorithms are designed for networks with only one type of relation between nodes but neglect the fact that many complex systems, including transportation and social networks, consisting of multi-modalities of interactions that correspond to different nature of interactions and dynamics that can be modeled as multiplex network, where different types of relation are represented in different layers. This paper proposes a Multi-Relations-aware Graph Neural Network (MRGNN) framework to learn effective node representations for multiplex networks and make more accurate link predictions, especially for weak ties. Specifically, our model utilizes an intra-layer node-level feature propagation process and an inter-layer representation merge process, which applies a simple yet effective logistic or semantic attention voting mechanism to adaptively aggregate information from different layers. Extensive experiments on four diversified multiplex networks show that MRGNN outperforms the state-of-the-art multiplex link prediction algorithms on overall prediction accuracy, and works pretty well on forecasting weak ties

cs.SI↗

arXiv2024

An Improved Strategy for Blood Glucose Control Using Multi-Step Deep Reinforcement Learning

Blood Glucose (BG) control involves keeping an individual's BG within a healthy range through extracorporeal insulin injections is an important task for people with type 1 diabetes. However,traditional patient self-management is cumbersome and risky. Recent research has been devoted to exploring individualized and automated BG control approaches, among which Deep Reinforcement Learning (DRL) shows potential as an emerging approach. In this paper, we use an exponential decay model of drug concentration to convert the formalization of the BG control problem, which takes into account the delay and prolongedness of drug effects, from a PAE-POMDP (Prolonged Action Effect-Partially Observable Markov Decision Process) to a MDP, and we propose a novel multi-step DRL-based algorithm to solve the problem. The Prioritized Experience Replay (PER) sampling method is also used in it. Compared to single-step bootstrapped updates, multi-step learning is more efficient and reduces the influence from biasing targets. Our proposed method converges faster and achieves higher cumulative rewards compared to the benchmark in the same training environment, and improves the time-in-range (TIR), the percentage of time the patient's BG is within the target range, in the evaluation phase. Our work validates the effectiveness of multi-step reinforcement learning in BG control, which may help to explore the optimal glycemic control measure and improve the survival of diabetic patients.

cs.AI↗

arXiv2024

Improving link prediction accuracy of network embedding algorithms via rich node attribute information

Complex networks are widely used to represent an abundance of real-world relations ranging from social networks to brain networks. Inferring missing links or predicting future ones based on the currently observed network is known as the link prediction task.Recent network embedding based link prediction algorithms have demonstrated ground-breaking performance on link prediction accuracy. Those algorithms usually apply node attributes as the initial feature input to accelerate the convergence speed during the training process. However, they do not take full advantage of node feature information. In this paper,besides applying feature attributes as the initial input, we make better utilization of node attribute information by building attributable networks and plugging attributable networks into some typical link prediction algorithms and naming this algorithm Attributive Graph Enhanced Embedding (AGEE). AGEE is able to automatically learn the weighting trades-off between the structure and the attributive networks. Numerical experiments show that AGEE can improve the link prediction accuracy by around 3% compared with link prediction framework SEAL, Variational Graph AutoEncoder (VGAE), and Node2vec.

cs.SI↗

arXiv2023

Interactive Visual Task Learning for Robots

We present a framework for robots to learn novel visual concepts and tasks via in-situ linguistic interactions with human users. Previous approaches have either used large pre-trained visual models to infer novel objects zero-shot, or added novel concepts along with their attributes and representations to a concept hierarchy. We extend the approaches that focus on learning visual concept hierarchies by enabling them to learn novel concepts and solve unseen robotics tasks with them. To enable a visual concept learner to solve robotics tasks one-shot, we developed two distinct techniques. Firstly, we propose a novel approach, Hi-Viscont(HIerarchical VISual CONcept learner for Task), which augments information of a novel concept to its parent nodes within a concept hierarchy. This information propagation allows all concepts in a hierarchy to update as novel concepts are taught in a continual learning setting. Secondly, we represent a visual task as a scene graph with language annotations, allowing us to create novel permutations of a demonstrated task zero-shot in-situ. We present two sets of results. Firstly, we compare Hi-Viscont with the baseline model (FALCON) on visual question answering(VQA) in three domains. While being comparable to the baseline model on leaf level concepts, Hi-Viscont achieves an improvement of over 9% on non-leaf concepts on average. We compare our model's performance against the baseline FALCON model. Our framework achieves 33% improvements in success rate metric, and 19% improvements in the object level accuracy compared to the baseline model. With both of these results we demonstrate the ability of our model to learn tasks and concepts in a continual learning setting on the robot.

cs.RO↗

arXiv2022

An Empirical Study on Finding Spans

We present an empirical study on methods for span finding, the selection of consecutive tokens in text for some downstream tasks. We focus on approaches that can be employed in training end-to-end information extraction systems, and find there is no definitive solution without considering task properties, and provide our observations to help with future design choices: 1) a tagging approach often yields higher precision while span enumeration and boundary prediction provide higher recall; 2) span type information can benefit a boundary prediction approach; 3) additional contextualization does not help span finding in most cases.

cs.CL↗

arXiv2022

Iterative Document-level Information Extraction via Imitation Learning

We present a novel iterative extraction model, IterX, for extracting complex relations, or templates (i.e., N-tuples representing a mapping from named slots to spans of text) within a document. Documents may feature zero or more instances of a template of any given type, and the task of template extraction entails identifying the templates in a document and extracting each template's slot values. Our imitation learning approach casts the problem as a Markov decision process (MDP), and relieves the need to use predefined template orders to train an extractor. It leads to state-of-the-art results on two established benchmarks -- 4-ary relation extraction on SciREX and template extraction on MUC-4 -- as well as a strong baseline on the new BETTER Granular task.

cs.CL↗

arXiv2020

Assessing the attraction of cities on venture capital from a scaling law perspective

Cities are centers for the integration of capital and incubators of invention, and attracting venture capital (VC) is of great importance for cities to advance in innovative technology and business models towards a sustainable and prosperous future. Yet we still lack a quantitative understanding of the relationship between urban characteristics and VC activities. In this paper, we find a clear nonlinear scaling relationship between VC activities and the urban population of Chinese cities. In such nonlinear systems, the widely applied linear per capita indicators would be either biased to larger cities or smaller cities depends on whether it is superlinear or sublinear, while the residual of cities relative to the prediction of scaling law is a more objective and scale-invariant metric. %(i.e., independent of the city size). Such a metric can distinguish the effects of local dynamics and scaled growth induced by the change of population size. The spatiotemporal evolution of such metrics on VC activities reveals three distinct groups of cities, two of which stand out with increasing and decreasing trends, respectively. And the taxonomy results together with spatial analysis also signify different development modes between large urban agglomeration regions. Besides, we notice the evolution of scaling exponents on VC activities are of much larger fluctuations than on socioeconomic output of cities, and a conceptual model that focuses on the growth dynamics of different sized cities can well explain it, which we assume would be general to other scenarios.

physics.soc-ph↗

arXiv2020

Principled approach to the selection of the embedding dimension of networks

Network embedding is a general-purpose machine learning technique that encodes network structure in vector spaces with tunable dimension. Choosing an appropriate embedding dimension -- small enough to be efficient and large enough to be effective -- is challenging but necessary to generate embeddings applicable to a multitude of tasks. Existing strategies for the selection of the embedding dimension rely on performance maximization in downstream tasks. Here, we propose a principled method such that all structural information of a network is parsimoniously encoded. The method is validated on various embedding algorithms and a large corpus of real-world networks. The embedding dimension selected by our method in real-world networks suggest that efficient encoding in low-dimensional spaces is usually possible.

physics.soc-ph↗

arXiv2019

Link Prediction via Graph Attention Network

Link prediction aims to infer missing links or predicting the future ones based on currently observed partial networks, it is a fundamental problem in network science with tremendous real-world applications. However, conventional link prediction approaches neither have high prediction accuracy nor being capable of revealing the hidden information behind links. To address this problem, we generalize the latest techniques in deep learning on graphs and present a new link prediction model - DeepLinker. Instead of learning node representation with the node label information, DeepLinker uses the links as supervised information. Experiments on five graphs show that DeepLinker can not only achieve the state-of-the-art link prediction accuracy, but also acquire the efficient node representations and node centrality ranking as the byproducts. Although the representations are obtained without any supervised node label information, they still perform well on node ranking and node classification tasks.

cs.SI↗

arXiv2018

Exploring Small-World Network with an Elite-Clique: Bringing Embeddedness Theory into the Dynamic Evolution of a Venture Capital Network

This paper uses a network dynamics model to explain the formation of a small-world network with an elite-clique. This network is a small-world network with an elite-clique at its center in which elites are also the centers of many small groups. These leaders also act as bridges between different small groups. Network dynamics are an important research topic due to their ability to explain the evolution of network structures. In this paper, a Chinese Venture Capital (VC) network was coded from joint investments between VC firms and then analyzed to uncover its network properties and factors that influence its evolution. We first built a random graph model to control for factors such as network scale, network growth, investment frequency and syndication tendency. Then we added a partner-selection mechanism and used two theories to analyze the formation of network structure: relational embeddedness and structural embeddedness. After that, we ran simulations and compared the three models with the actual Chinese VC network. To do this we computed the elite-clique's EI index, degree distribution, clustering coefficient distribution and motifs. Results show that adding embeddedness theories significantly improved the network dynamic model's predictive power, and help us uncover the mechanisms that affect the formation of a small-world industrial network with an elite-clique at its center.

cs.SI↗

arXiv2017

The Hidden Flow Structure and Metric Space of Network Embedding Algorithms Based on Random Walks

Network embedding which encodes all vertices in a network as a set of numerical vectors in accordance with it's local and global structures, has drawn widespread attention. Network embedding not only learns significant features of a network, such as the clustering and linking prediction but also learns the latent vector representation of the nodes which provides theoretical support for a variety of applications, such as visualization, node classification, and recommendation. As the latest progress of the research, several algorithms based on random walks have been devised. Although their high scores for learning efficiency and accuracy have drawn much attention, there is still a lack of theoretical explanation, and the transparency of the algorithms has been doubted. Here, we propose an approach based on the open-flow network model to reveal the underlying flow structure and its hidden metric space of different random walk strategies on networks. We show that the essence of embedding based on random walks is the latent metric structure defined on the open-flow network. This not only deepens our understanding of random walk based embedding algorithms but also helps in finding new potential applications in embedding.

physics.soc-ph↗