We consider the role of non-localities in speed-density data used to fit fundamental diagrams from vehicle trajectories. We demonstrate that the use of anticipated densities results in a clear classification of speed-density data into stationary and non-stationary points, namely, acceleration and deceleration regimes and their separating boundary. The separating boundary represents a locus of stationary traffic states, i.e., the fundamental diagram. To fit fundamental diagrams, we develop an enhanced cross entropy minimization method that honors equilibrium traffic physics. We illustrate the effectiveness of our proposed approach by comparing it with the traditional approach that uses local speed-density states and least squares estimation. Our experiments show that the separating boundary in our approach is invariant to varying trajectory samples within the same spatio-temporal region, providing further evidence that the separating boundary is indeed a locus of stationary traffic states.
相關內容
Robotic manipulation can greatly benefit from the data efficiency, robustness, and predictability of model-based methods if robots can quickly generate models of novel objects they encounter. This is especially difficult when effects like complex joint friction lack clear first-principles models and are usually ignored by physics simulators. Further, numerically-stiff contact dynamics can make common model-building approaches struggle. We propose a method to simultaneously learn contact and continuous dynamics of a novel, possibly multi-link object by observing its motion through contact-rich trajectories. We formulate a system identification process with a loss that infers unmeasured contact forces, penalizing their violation of physical constraints and laws of motion given current model parameters. Our loss is unlike prediction-based losses used in differentiable simulation. Using a new dataset of real articulated object trajectories and an existing cube toss dataset, our method outperforms differentiable simulation and end-to-end alternatives with more data efficiency. See our project page for code, datasets, and media: //sites.google.com/view/continuous-contact-nets/home
Explainable recommender systems (RS) have traditionally followed a one-size-fits-all approach, delivering the same explanation level of detail to each user, without considering their individual needs and goals. Further, explanations in RS have so far been presented mostly in a static and non-interactive manner. To fill these research gaps, we aim in this paper to adopt a user-centered, interactive explanation model that provides explanations with different levels of detail and empowers users to interact with, control, and personalize the explanations based on their needs and preferences. We followed a user-centered approach to design interactive explanations with three levels of detail (basic, intermediate, and advanced) and implemented them in the transparent Recommendation and Interest Modeling Application (RIMA). We conducted a qualitative user study (N=14) to investigate the impact of providing interactive explanations with varying level of details on the users' perception of the explainable RS. Our study showed qualitative evidence that fostering interaction and giving users control in deciding which explanation they would like to see can meet the demands of users with different needs, preferences, and goals, and consequently can have positive effects on different crucial aspects in explainable recommendation, including transparency, trust, satisfaction, and user experience.
The growing demand for electric vehicles requires the development of automated car charging methods. At the moment, the process of charging an electric car is completely manual, and that requires physical effort to accomplish the task, which is not suitable for people with disabilities. Typically, the effort in the research is focused on detecting the position and orientation of the socket, which resulted in a relatively high accuracy, $\pm 5 \: mm $ and $\pm 10^o$. However, this accuracy is not enough to complete the charging process. In this work, we focus on designing a novel methodology for robust robotic plug-in and plug-out based on human haptics, to overcome the error in the position and orientation of the socket. Participants were invited to perform the charging task, and their cognitive capabilities were recognized by measuring the applied forces along with the movement of the charger. Three controllers were designed based on impedance control to mimic the human patterns of charging an electric car. The recorded data from humans were used to calibrate the parameters of the impedance controllers: inertia $M_d$, damping $D_d$, and stiffness $K_d$. A robotic validation was performed, where the designed controllers were applied to the robot UR10. Using the proposed controllers and the human kinesthetic data, it was possible to successfully automate the operation of charging an electric car.
Every major technical invention resurfaces the dual-use dilemma -- the new technology has the potential to be used for good as well as for harm. Generative AI (GenAI) techniques, such as large language models (LLMs) and diffusion models, have shown remarkable capabilities (e.g., in-context learning, code-completion, and text-to-image generation and editing). However, GenAI can be used just as well by attackers to generate new attacks and increase the velocity and efficacy of existing attacks. This paper reports the findings of a workshop held at Google (co-organized by Stanford University and the University of Wisconsin-Madison) on the dual-use dilemma posed by GenAI. This paper is not meant to be comprehensive, but is rather an attempt to synthesize some of the interesting findings from the workshop. We discuss short-term and long-term goals for the community on this topic. We hope this paper provides both a launching point for a discussion on this important topic as well as interesting problems that the research community can work to address.
To benefit from the abundance of data and the insights it brings data processing pipelines are being used in many areas of research and development in both industry and academia. One approach to automating data processing pipelines is the workflow technology, as it also supports collaborative, trial-and-error experimentation with the pipeline architecture in different application domains. In addition to the necessary flexibility that such pipelines need to possess, in collaborative settings cross-organisational interactions are plagued by lack of trust. While capturing provenance information related to the pipeline execution and the processed data is a first step towards enabling trusted collaborations, the current solutions do not allow for provenance of the change in the processing pipelines, where the subject of change can be made on any aspect of the workflow implementing the pipeline and on the data used while the pipeline is being executed. Therefore in this work we provide a solution architecture and a proof of concept implementation of a service, called Provenance Holder, which enable provenance of collaborative, adaptive data processing pipelines in a trusted manner. We also contribute a definition of a set of properties of such a service and identify future research directions.
Data-driven predictions are often perceived as inaccurate in hindsight due to behavioral responses. We consider the role of interface design choices on how individuals respond to predictions presented on a shared information display in a strategic setting. We introduce a novel staged experimental design to investigate the effects of interface design features, such as the visualization of prediction uncertainty and prediction error, within a repeated congestion game. In this game, participants assume the role of taxi drivers and use a shared information display to decide where to search for their next ride. Our experimental design endows agents with varying level-$k$ depths of thinking, allowing some agents to possess greater sophistication in anticipating the decisions of others using the same information display. Through several large pre-registered experiments, we identify trade-offs between displays that are optimal for individual decisions and those that best serve the collective social welfare of the system. Additionally, we note that the influence of display characteristics varies based on an agent's strategic sophistication. We observe that design choices promoting individual-level decision-making can lead to suboptimal system outcomes, as manifested by a lower realization of potential social welfare. However, this decline in social welfare is offset by a slight reduction in distribution shift, narrowing the gap between predicted and realized system outcomes. This may enhance the perceived reliability and trustworthiness of the information display post hoc. Our findings pave the way for new research questions concerning the design of effective prediction interfaces in strategic environments.
The growing demand for electric vehicles requires the development of automated car charging methods. At the moment, the process of charging an electric car is completely manual, and that requires physical effort to accomplish the task, which is not suitable for people with disabilities. Typically, the effort in the automation of the charging task research is focused on detecting the position and orientation of the socket, which resulted in a relatively high accuracy, 5 mm, and 10 degrees. However, this accuracy is not enough to complete the charging process. In this work, we focus on designing a novel methodology for robust robotic plug-in and plug-out based on human haptics to overcome the error in the orientation of the socket. Participants were invited to perform the charging task, and their cognitive capabilities were recognized by measuring the applied forces along with the movements of the charger. Eventually, an algorithm was developed based on the human's best strategies to be applied to a robotic arm.
The dominating NLP paradigm of training a strong neural predictor to perform one task on a specific dataset has led to state-of-the-art performance in a variety of applications (eg. sentiment classification, span-prediction based question answering or machine translation). However, it builds upon the assumption that the data distribution is stationary, ie. that the data is sampled from a fixed distribution both at training and test time. This way of training is inconsistent with how we as humans are able to learn from and operate within a constantly changing stream of information. Moreover, it is ill-adapted to real-world use cases where the data distribution is expected to shift over the course of a model's lifetime. The first goal of this thesis is to characterize the different forms this shift can take in the context of natural language processing, and propose benchmarks and evaluation metrics to measure its effect on current deep learning architectures. We then proceed to take steps to mitigate the effect of distributional shift on NLP models. To this end, we develop methods based on parametric reformulations of the distributionally robust optimization framework. Empirically, we demonstrate that these approaches yield more robust models as demonstrated on a selection of realistic problems. In the third and final part of this thesis, we explore ways of efficiently adapting existing models to new domains or tasks. Our contribution to this topic takes inspiration from information geometry to derive a new gradient update rule which alleviate catastrophic forgetting issues during adaptation.
AI is undergoing a paradigm shift with the rise of models (e.g., BERT, DALL-E, GPT-3) that are trained on broad data at scale and are adaptable to a wide range of downstream tasks. We call these models foundation models to underscore their critically central yet incomplete character. This report provides a thorough account of the opportunities and risks of foundation models, ranging from their capabilities (e.g., language, vision, robotics, reasoning, human interaction) and technical principles(e.g., model architectures, training procedures, data, systems, security, evaluation, theory) to their applications (e.g., law, healthcare, education) and societal impact (e.g., inequity, misuse, economic and environmental impact, legal and ethical considerations). Though foundation models are based on standard deep learning and transfer learning, their scale results in new emergent capabilities,and their effectiveness across so many tasks incentivizes homogenization. Homogenization provides powerful leverage but demands caution, as the defects of the foundation model are inherited by all the adapted models downstream. Despite the impending widespread deployment of foundation models, we currently lack a clear understanding of how they work, when they fail, and what they are even capable of due to their emergent properties. To tackle these questions, we believe much of the critical research on foundation models will require deep interdisciplinary collaboration commensurate with their fundamentally sociotechnical nature.
Object detection typically assumes that training and test data are drawn from an identical distribution, which, however, does not always hold in practice. Such a distribution mismatch will lead to a significant performance drop. In this work, we aim to improve the cross-domain robustness of object detection. We tackle the domain shift on two levels: 1) the image-level shift, such as image style, illumination, etc, and 2) the instance-level shift, such as object appearance, size, etc. We build our approach based on the recent state-of-the-art Faster R-CNN model, and design two domain adaptation components, on image level and instance level, to reduce the domain discrepancy. The two domain adaptation components are based on H-divergence theory, and are implemented by learning a domain classifier in adversarial training manner. The domain classifiers on different levels are further reinforced with a consistency regularization to learn a domain-invariant region proposal network (RPN) in the Faster R-CNN model. We evaluate our newly proposed approach using multiple datasets including Cityscapes, KITTI, SIM10K, etc. The results demonstrate the effectiveness of our proposed approach for robust object detection in various domain shift scenarios.
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191