arXiv subjects

Lu Feng

Publications and source records attributed to Lu Feng.

At least 55 records · Page 3Linked to original sources

arXiv2024

Safe POMDP Online Planning among Dynamic Agents via Adaptive Conformal Prediction

Online planning for partially observable Markov decision processes (POMDPs) provides efficient techniques for robot decision-making under uncertainty. However, existing methods fall short of preventing safety violations in dynamic environments. This work presents a novel safe POMDP online planning approach that maximizes expected returns while providing probabilistic safety guarantees amidst environments populated by multiple dynamic agents. Our approach utilizes data-driven trajectory prediction models of dynamic agents and applies Adaptive Conformal Prediction (ACP) to quantify the uncertainties in these predictions. Leveraging the obtained ACP-based trajectory predictions, our approach constructs safety shields on-the-fly to prevent unsafe actions within POMDP online planning. Through experimental evaluation in various dynamic environments using real-world pedestrian trajectory data, the proposed approach has been shown to effectively maintain probabilistic safety guarantees while accommodating up to hundreds of dynamic agents.

cs.RO↗

arXiv2024

Prospects for weighing neutrinos in interacting dark energy models using joint observations of gravitational waves and $γ$-ray bursts

Cosmological observations can be used to weigh neutrinos, but this method is model-dependent, with results relying on the cosmological model considered. If we consider interactions between dark energy and dark matter, the neutrino mass constraints differ from those derived under the standard model. On the contrary, gravitational wave (GW) standard siren observations can measure absolute cosmological distances, helping to break parameter degeneracies inherent in traditional cosmological observations, thereby improving constraints on neutrino mass. This paper examines the constraints on neutrino mass within interacting dark energy (IDE) models and explores how future GW standard siren observations could enhance these results. For multi-messenger GW observations, we consider the joint observations of binary neutron star mergers by third-generation ground-based GW detectors and short $γ$-ray burst observations by missions similar to the THESEUS satellite project. Using current cosmological observations (CMB+BAO+SN), we obtain an upper limit on the neutrino mass in the IDE models of 0.15 (or 0.16) eV. With the inclusion of GW data, the upper limit on the neutrino mass improves to 0.14 eV. This indicates that in the context of IDE models, the improvement in neutrino mass constraints from GW observations is relatively limited. However, GW observations significantly enhance the constraints on other cosmological parameters, such as matter density parameter, the Hubble constant, and coupling strength between dark energy and dark matter.

astro-ph.CO↗

arXiv2024

DP-RuL: Differentially-Private Rule Learning for Clinical Decision Support Systems

Serious privacy concerns arise with the use of patient data in rule-based clinical decision support systems (CDSS). The goal of a privacy-preserving CDSS is to learn a population ruleset from individual clients' local rulesets, while protecting the potentially sensitive information contained in the rulesets. We present the first work focused on this problem and develop a framework for learning population rulesets with local differential privacy (LDP), suitable for use within a distributed CDSS and other distributed settings. Our rule discovery protocol uses a Monte-Carlo Tree Search (MCTS) method integrated with LDP to search a rule grammar in a structured way and find rule structures clients are likely to have. Randomized response queries are sent to clients to determine promising paths to search within the rule grammar. In addition, we introduce an adaptive budget allocation method which dynamically determines how much privacy loss budget to use at each query, resulting in better privacy-utility trade-offs. We evaluate our approach using three clinical datasets and find that we are able to learn population rulesets with high coverage (breadth of rules) and clinical utility even at low privacy loss budgets.

cs.CR↗

arXiv2024

Distributional Probabilistic Model Checking

Probabilistic model checking can provide formal guarantees on the behavior of stochastic models relating to a wide range of quantitative properties, such as runtime, energy consumption or cost. But decision making is typically with respect to the expected value of these quantities, which can mask important aspects of the full probability distribution such as the possibility of high-risk, low-probability events or multimodalities. We propose a distributional extension of probabilistic model checking, applicable to discrete-time Markov chains (DTMCs) and Markov decision processes (MDPs). We formulate distributional queries, which can reason about a variety of distributional measures, such as variance, value-at-risk or conditional value-at-risk, for the accumulation of reward until a co-safe linear temporal logic formula is satisfied. For DTMCs, we propose a method to compute the full distribution to an arbitrary level of precision, based on a graph analysis and forward analysis of the model. For MDPs, we approximate the optimal policy with respect to expected value or conditional value-at-risk using distributional value iteration. We implement our techniques and investigate their performance and scalability across a range of benchmark models. Experimental results demonstrate that our techniques can be successfully applied to check various distributional properties of large probabilistic models.

cs.LO↗

arXiv2024

Safe POMDP Online Planning via Shielding

Partially observable Markov decision processes (POMDPs) have been widely used in many robotic applications for sequential decision-making under uncertainty. POMDP online planning algorithms such as Partially Observable Monte-Carlo Planning (POMCP) can solve very large POMDPs with the goal of maximizing the expected return. But the resulting policies cannot provide safety guarantees which are imperative for real-world safety-critical tasks (e.g., autonomous driving). In this work, we consider safety requirements represented as almost-sure reach-avoid specifications (i.e., the probability to reach a set of goal states is one and the probability to reach a set of unsafe states is zero). We compute shields that restrict unsafe actions which would violate the almost-sure reach-avoid specifications. We then integrate these shields into the POMCP algorithm for safe POMDP online planning. We propose four distinct shielding methods, differing in how the shields are computed and integrated, including factored variants designed to improve scalability. Experimental results on a set of benchmark domains demonstrate that the proposed shielding methods successfully guarantee safety (unlike the baseline POMCP without shielding) on large POMDPs, with negligible impact on the runtime for online planning.

cs.AI↗

arXiv2023

GlucoSynth: Generating Differentially-Private Synthetic Glucose Traces

We focus on the problem of generating high-quality, private synthetic glucose traces, a task generalizable to many other time series sources. Existing methods for time series data synthesis, such as those using Generative Adversarial Networks (GANs), are not able to capture the innate characteristics of glucose data and cannot provide any formal privacy guarantees without severely degrading the utility of the synthetic data. In this paper we present GlucoSynth, a novel privacy-preserving GAN framework to generate synthetic glucose traces. The core intuition behind our approach is to conserve relationships amongst motifs (glucose events) within the traces, in addition to temporal dynamics. Our framework incorporates differential privacy mechanisms to provide strong formal privacy guarantees. We provide a comprehensive evaluation on the real-world utility of the data using 1.2 million glucose traces; GlucoSynth outperforms all previous methods in its ability to generate high-quality synthetic glucose traces with strong privacy guarantees.

cs.LG↗

arXiv2023

Trust-Aware Motion Planning for Human-Robot Collaboration under Distribution Temporal Logic Specifications

Recent work has considered trust-aware decision making for human-robot collaboration (HRC) with a focus on model learning. In this paper, we are interested in enabling the HRC system to complete complex tasks specified using temporal logic that involve human trust. Since human trust in robots is not observable, we adopt the widely used partially observable Markov decision process (POMDP) framework for modelling the interactions between humans and robots. To specify the desired behaviour, we propose to use syntactically co-safe linear distribution temporal logic (scLDTL), a logic that is defined over predicates of states as well as belief states of partially observable systems. The incorporation of belief predicates in scLDTL enhances its expressiveness while simultaneously introducing added complexity. This also presents a new challenge as the belief predicates must be evaluated over the continuous (infinite) belief space. To address this challenge, we present an algorithm for solving the optimal policy synthesis problem. First, we enhance the belief MDP (derived by reformulating the POMDP) with a probabilistic labelling function. Then a product belief MDP is constructed between the probabilistically labelled belief MDP and the automaton translation of the scLDTL formula. Finally, we show that the optimal policy can be obtained by leveraging existing point-based value iteration algorithms with essential modifications. Human subject experiments with 21 participants on a driving simulator demonstrate the effectiveness of the proposed approach.

cs.RO↗

arXiv2023

CARNA: Characterizing Advanced heart failure Risk and hemodyNAmic phenotypes using learned multi-valued decision diagrams

Early identification of high risk heart failure (HF) patients is key to timely allocation of life-saving therapies. Hemodynamic assessments can facilitate risk stratification and enhance understanding of HF trajectories. However, risk assessment for HF is a complex, multi-faceted decision-making process that can be challenging. Previous risk models for HF do not integrate invasive hemodynamics or support missing data, and use statistical methods prone to bias or machine learning methods that are not interpretable. To address these limitations, this paper presents CARNA, a hemodynamic risk stratification and phenotyping framework for advanced HF that takes advantage of the explainability and expressivity of machine learned Multi-Valued Decision Diagrams (MVDDs). This interpretable framework learns risk scores that predict the probability of patient outcomes, and outputs descriptive patient phenotypes (sets of features and thresholds) that characterize each predicted risk score. CARNA incorporates invasive hemodynamics and can make predictions on missing data. The CARNA models were trained and validated using a total of five advanced HF patient cohorts collected from previous trials, and compared with six established HF risk scores and three traditional ML risk models. CARNA provides robust risk stratification, outperforming all previous benchmarks. Although focused on advanced HF, the CARNA framework is general purpose and can be used to learn risk stratifications for other diseases and medical applications.

cs.LG↗

arXiv2023

Explainable Multi-Agent Reinforcement Learning for Temporal Queries

As multi-agent reinforcement learning (MARL) systems are increasingly deployed throughout society, it is imperative yet challenging for users to understand the emergent behaviors of MARL agents in complex environments. This work presents an approach for generating policy-level contrastive explanations for MARL to answer a temporal user query, which specifies a sequence of tasks completed by agents with possible cooperation. The proposed approach encodes the temporal query as a PCTL logic formula and checks if the query is feasible under a given MARL policy via probabilistic model checking. Such explanations can help reconcile discrepancies between the actual and anticipated multi-agent behaviors. The proposed approach also generates correct and complete explanations to pinpoint reasons that make a user query infeasible. We have successfully applied the proposed approach to four benchmark MARL domains (up to 9 agents in one domain). Moreover, the results of a user study show that the generated explanations significantly improve user performance and satisfaction.

cs.AI↗

arXiv2023

First sodium laser guide star asterism launching platform in China on 1.8m telescope at Gaomeigu Observatory

The application of sodium laser guide star is the key difference between modern adaptive optics system and traditional adaptive optics system. Especially in system like multi-conjugate adaptive optics, sodium laser guide star asterism which is formed by several laser guide stars in certain pattern is required to probe more atmospheric turbulence in different directions. To achieve this, a sodium laser guide star asterism launching platform is required. In this paper, we will introduce the sodium laser guide star asterism launching platform built and tested on the 1.8m telescope of the Gaomeigu Observatory. The platform has two functions: one is to compare the performance of sodium laser guide stars generated by different lasers at the same place; the other is to generate sodium laser guide star asterism with adjustable shape. The field test results at the beginning of 2021 verify the important role of the platform, which is also the first time to realize sodium laser guide star asterism in China.

astro-ph.IM↗

arXiv2022

Towards Developing Safety Assurance Cases for Learning-Enabled Medical Cyber-Physical Systems

Machine Learning (ML) technologies have been increasingly adopted in Medical Cyber-Physical Systems (MCPS) to enable smart healthcare. Assuring the safety and effectiveness of learning-enabled MCPS is challenging, as such systems must account for diverse patient profiles and physiological dynamics and handle operational uncertainties. In this paper, we develop a safety assurance case for ML controllers in learning-enabled MCPS, with an emphasis on establishing confidence in the ML-based predictions. We present the safety assurance case in detail for Artificial Pancreas Systems (APS) as a representative application of learning-enabled MCPS, and provide a detailed analysis by implementing a deep neural network for the prediction in APS. We check the sufficiency of the ML data and analyze the correctness of the ML-based prediction using formal verification. Finally, we outline open research problems based on our experience in this paper.

cs.LG↗

arXiv2022

Enjoy the Ride Consciously with CAWA: Context-Aware Advisory Warnings for Automated Driving

In conditionally automated driving, drivers decoupled from driving while immersed in non-driving-related tasks (NDRTs) could potentially either miss the system-initiated takeover request (TOR) or a sudden TOR may startle them. To better prepare drivers for a safer takeover in an emergency, we propose novel context-aware advisory warnings (CAWA) for automated driving to gently inform drivers. This will help them stay vigilant while engaging in NDRTs. The key innovation is that CAWA adapts warning modalities according to the context of NDRTs. We conducted a user study to investigate the effectiveness of CAWA. The study results show that CAWA has statistically significant effects on safer takeover behavior, improved driver situational awareness, less attention demand, and more positive user feedback, compared with uniformly distributed speech-based warnings across all NDRTs.

cs.HC↗

arXiv2022

Planning for Automated Vehicles with Human Trust

Recent work has considered personalized route planning based on user profiles, but none of it accounts for human trust. We argue that human trust is an important factor to consider when planning routes for automated vehicles. This paper presents a trust-based route planning approach for automated vehicles. We formalize the human-vehicle interaction as a partially observable Markov decision process (POMDP) and model trust as a partially observable state variable of the POMDP, representing the human's hidden mental state. We build data-driven models of human trust dynamics and takeover decisions, which are incorporated in the POMDP framework, using data collected from an online user study with 100 participants on the Amazon Mechanical Turk platform. We compute optimal routes for automated vehicles by solving optimal policies in the POMDP planning, and evaluate the resulting routes via human subject experiments with 22 participants on a driving simulator. The experimental results show that participants taking the trust-based route generally reported more positive responses in the after-driving survey than those taking the baseline (trust-free) route. In addition, we analyze the trade-offs between multiple planning objectives (e.g., trust, distance, energy consumption) via multi-objective optimization of the POMDP. We also identify a set of open issues and implications for real-world deployment of the proposed approach in automated vehicles.

cs.HC↗

arXiv2022

A Study on Learning and Simulating Personalized Car-Following Driving Style

Automated vehicles are gradually entering people's daily life to provide a comfortable driving experience for the users. The generic and user-agnostic automated vehicles have limited ability to accommodate the different driving styles of different users. This limitation not only impacts users' satisfaction but also causes safety concerns. Learning from user demonstrations can provide direct insights regarding users' driving preferences. However, it is difficult to understand a driver's preference with limited data. In this study, we use a model-free inverse reinforcement learning method to study drivers' characteristics in the car-following scenario from a naturalistic driving dataset, and show this method is capable of representing users' preferences with reward functions. In order to predict the driving styles for drivers with limited data, we apply Gaussian Mixture Models and compute the similarity of a specific driver to the clusters of drivers. We design a personalized adaptive cruise control (P-ACC) system through a partially observable Markov decision process (POMDP) model. The reward function with the model to mimic drivers' driving style is integrated, with a constraint on the relative distance to ensure driving safety. Prediction of the driving styles achieves 85.7% accuracy with the data of less than 10 car-following events. The model-based experimental driving trajectories demonstrate that the P-ACC system can provide a personalized driving experience.

cs.HC↗

arXiv2022

Logic-based Reward Shaping for Multi-Agent Reinforcement Learning

Reinforcement learning (RL) relies heavily on exploration to learn from its environment and maximize observed rewards. Therefore, it is essential to design a reward function that guarantees optimal learning from the received experience. Previous work has combined automata and logic based reward shaping with environment assumptions to provide an automatic mechanism to synthesize the reward function based on the task. However, there is limited work on how to expand logic-based reward shaping to Multi-Agent Reinforcement Learning (MARL). The environment will need to consider the joint state in order to keep track of other agents if the task requires cooperation, thus suffering from the curse of dimensionality with respect to the number of agents. This project explores how logic-based reward shaping for MARL can be designed for different scenarios and tasks. We present a novel method for semi-centralized logic-based MARL reward shaping that is scalable in the number of agents and evaluate it in multiple scenarios.

cs.AI↗

arXiv2022

Toward Policy Explanations for Multi-Agent Reinforcement Learning

Advances in multi-agent reinforcement learning (MARL) enable sequential decision making for a range of exciting multi-agent applications such as cooperative AI and autonomous driving. Explaining agent decisions is crucial for improving system transparency, increasing user satisfaction, and facilitating human-agent collaboration. However, existing works on explainable reinforcement learning mostly focus on the single-agent setting and are not suitable for addressing challenges posed by multi-agent environments. We present novel methods to generate two types of policy explanations for MARL: (i) policy summarization about the agent cooperation and task sequence, and (ii) language explanations to answer queries about agent behavior. Experimental results on three MARL domains demonstrate the scalability of our methods. A user study shows that the generated explanations significantly improve user performance and increase subjective ratings on metrics such as user satisfaction.

cs.AI↗

arXiv2022

Photometric redshifts and Galaxy Clusters for DES DR2, DESI DR9, and HSC-SSP PDR3 Data

Photometric redshift (photo-z) is a fundamental parameter for multi-wavelength photometric surveys, while galaxy clusters are important cosmological probers and ideal objects for exploring the dense environmental impact on galaxy evolution. We extend our previous work on estimating photo-z and detecting galaxy clusters to the latest data releases of the Dark Energy Spectroscopic Instrument (DESI) imaging surveys, Dark Energy Survey (DES), and Hyper Suprime-Cam Subaru Strategic Program (HSC-SSP) imaging surveys and make corresponding catalogs publicly available for more extensive scientific applications. The photo-z catalogs include accurate measurements of photo-z and stellar mass for about 320, 293, and 134 million galaxies with $r<23$, $i<24$, and $i<25$ in DESI DR9, DES DR2, and HSC-SSP PDR3 data, respectively. The photo-z accuracy is about 0.017, 0.024, and 0.029 and the general redshift coverage is $z<1$, $z<1.2$, and $z<1.6$, respectively for those three surveys. The uncertainties of the logarithmic stellar mass that is inferred from stellar population synthesis fitting is about 0.2 dex. With the above photo-z catalogs, galaxy clusters are detected using a fast cluster-finding algorithm. A total of 532,810, 86,963, and 36,566 galaxy clusters with the number of members larger than 10 are discovered for DESI, DES, and HSC-SSP, respectively. Their photo-z accuracy is at the level of 0.01. The total mass of our clusters are also estimated by using the calibration relations between the optical richness and the mass measurement from X-ray and radio observations. The photo-z and cluster catalogs are available at ScienceDB (https://www.doi.org/10.11922/sciencedb.o00069.00003) and PaperData Repository (https://doi.org/10.12149/101089).

astro-ph.GA↗

arXiv2022

Multi-Objective Controller Synthesis with Uncertain Human Preferences

Complex real-world applications of cyber-physical systems give rise to the need for multi-objective controller synthesis, which concerns the problem of computing an optimal controller subject to multiple (possibly conflicting) criteria. The relative importance of objectives is often specified by human decision-makers. However, there is inherent uncertainty in human preferences (e.g., due to artifacts resulting from different preference elicitation methods). In this paper, we formalize the notion of uncertain human preferences and present a novel approach that accounts for this uncertainty in the context of multi-objective controller synthesis for Markov decision processes (MDPs). Our approach is based on mixed-integer linear programming and synthesizes an optimally permissive multi-strategy that satisfies uncertain human preferences with respect to a multi-objective property. Experimental results on a range of large case studies show that the proposed approach is feasible and scalable across varying MDP model sizes and uncertainty levels of human preferences. Evaluation via an online user study also demonstrates the quality and benefits of the synthesized controllers.

cs.AI↗