David Garlan

Md Abir Hossen, Sonam Kharade, Bradley Schmerl et al.

Robotic systems have subsystems with a combinatorially large configuration space and hundreds or thousands of possible software and hardware configuration options interacting non-trivially. The configurable parameters are set to target specific objectives, but they can cause functional faults when incorrectly configured. Finding the root cause of such faults is challenging due to the exponentially large configuration space and the dependencies between the robot's configuration settings and performance. This paper proposes CaRE -- a method for diagnosing the root cause of functional faults through the lens of causality. CaRE abstracts the causal relationships between various configuration options and the robot's performance objectives by learning a causal structure and estimating the causal effects of options on robot performance indicators. We demonstrate CaRE's efficacy by finding the root cause of the observed functional faults and validating the diagnosed root cause by conducting experiments in both physical robots (Husky and Turtlebot 3) and in simulation (Gazebo). Furthermore, we demonstrate that the causal models learned from robots in simulation (e.g., Husky in Gazebo) are transferable to physical robots across different platforms (e.g., Husky and Turtlebot 3).

Pedro Mendes, Paolo Romano, David Garlan

This work focuses on the problem of hyper-parameter tuning (HPT) for robust (i.e., adversarially trained) models, shedding light on the new challenges and opportunities arising during the HPT process for robust models. To this end, we conduct an extensive experimental study based on 3 popular deep models, in which we explore exhaustively 9 (discretized) HPs, 2 fidelity dimensions, and 2 attack bounds, for a total of 19208 configurations (corresponding to 50 thousand GPU hours). Through this study, we show that the complexity of the HPT problem is further exacerbated in adversarial settings due to the need to independently tune the HPs used during standard and adversarial training: succeeding in doing so (i.e., adopting different HP settings in both phases) can lead to a reduction of up to 80% and 43% of the error for clean and adversarial inputs, respectively. On the other hand, we also identify new opportunities to reduce the cost of HPT for robust models. Specifically, we propose to leverage cheap adversarial training methods to obtain inexpensive, yet highly correlated, estimations of the quality achievable using state-of-the-art methods. We show that, by exploiting this novel idea in conjunction with a recent multi-fidelity optimizer (taKG), the efficiency of the HPT process can be enhanced by up to 2.1x.

Pedro Mendes, Paolo Romano, David Garlan

Reliable uncertainty estimation is critical for deploying neural networks (NNs) in real-world applications. While existing calibration techniques often rely on post-hoc adjustments or coarse-grained binning methods, they remain limited in scalability, differentiability, and generalization across domains. In this work, we introduce CLUE (Calibration via Learning Uncertainty-Error Alignment), a novel approach that explicitly aligns predicted uncertainty with observed error during training, grounded in the principle that well-calibrated models should produce uncertainty estimates that match their empirical loss. CLUE adopts a novel loss function that jointly optimizes predictive performance and calibration, using summary statistics of uncertainty and loss as proxies. The proposed method is fully differentiable, domain-agnostic, and compatible with standard training pipelines. Through extensive experiments on vision, regression, and language modeling tasks, including out-of-distribution and domain-shift scenarios, we demonstrate that CLUE achieves superior calibration quality and competitive predictive performance with respect to state-of-the-art approaches without imposing significant computational overhead.

Pedro Mendes, Paolo Romano, David Garlan

Neural networks are often overconfident about their predictions, which undermines their reliability and trustworthiness. In this work, we present a novel technique, named Error-Driven Uncertainty Aware Training (EUAT), which aims to enhance the ability of neural classifiers to estimate their uncertainty correctly, namely to be highly uncertain when they output inaccurate predictions and low uncertain when their output is accurate. The EUAT approach operates during the model's training phase by selectively employing two loss functions depending on whether the training examples are correctly or incorrectly predicted by the model. This allows for pursuing the twofold goal of i) minimizing model uncertainty for correctly predicted inputs and ii) maximizing uncertainty for mispredicted inputs, while preserving the model's misprediction rate. We evaluate EUAT using diverse neural models and datasets in the image recognition domains considering both non-adversarial and adversarial settings. The results show that EUAT outperforms existing approaches for uncertainty estimation (including other uncertainty-aware training techniques, calibration, ensembles, and DEUP) by providing uncertainty estimates that not only have higher quality when evaluated via statistical metrics (e.g., correlation with residuals) but also when employed to build binary classifiers that decide whether the model's output can be trusted or not and under distributional data shifts.

Md Abir Hossen, Sonam Kharade, Jason M. O'Kane et al.

Robotic systems are typically composed of various subsystems, such as localization and navigation, each encompassing numerous configurable components (e.g., selecting different planning algorithms). Once an algorithm has been selected for a component, its associated configuration options must be set to the appropriate values. Configuration options across the system stack interact non-trivially. Finding optimal configurations for highly configurable robots to achieve desired performance poses a significant challenge due to the interactions between configuration options across software and hardware that result in an exponentially large and complex configuration space. These challenges are further compounded by the need for transferability between different environments and robotic platforms. Data efficient optimization algorithms (e.g., Bayesian optimization) have been increasingly employed to automate the tuning of configurable parameters in cyber-physical systems. However, such optimization algorithms converge at later stages, often after exhausting the allocated budget (e.g., optimization steps, allotted time) and lacking transferability. This paper proposes CURE -- a method that identifies causally relevant configuration options, enabling the optimization process to operate in a reduced search space, thereby enabling faster optimization of robot performance. CURE abstracts the causal relationships between various configuration options and robot performance objectives by learning a causal model in the source (a low-cost environment such as the Gazebo simulator) and applying the learned knowledge to perform optimization in the target (e.g., Turtlebot 3 physical robot). We demonstrate the effectiveness and transferability of CURE by conducting experiments that involve varying degrees of deployment changes in both physical robots and simulation.

Pedro Mendes, Paolo Romano, David Garlan

Modern neural networks (NNs) often achieve high predictive accuracy but are poorly calibrated, producing overconfident predictions even when wrong. This miscalibration poses serious challenges in applications where reliable uncertainty estimates are critical. In this work, we investigate how human perceptual uncertainty compares to uncertainty estimated by NNs. Using three vision benchmarks annotated with both human disagreement and crowdsourced confidence, we assess the correlation between model-predicted uncertainty and human-perceived uncertainty. Our results show that current methods only weakly align with human intuition, with correlations varying significantly across tasks and uncertainty metrics. Notably, we find that incorporating human-derived soft labels into the training process can improve calibration without compromising accuracy. These findings reveal a persistent gap between model and human uncertainty and highlight the potential of leveraging human insights to guide the development of more trustworthy AI systems.

Changjian Zhang, Parv Kapoor, Eunsuk Kang et al.

Cyber-physical systems (CPS) with reinforcement learning (RL)-based controllers are increasingly being deployed in complex physical environments such as autonomous vehicles, the Internet-of-Things(IoT), and smart cities. An important property of a CPS is tolerance; i.e., its ability to function safely under possible disturbances and uncertainties in the actual operation. In this paper, we introduce a new, expressive notion of tolerance that describes how well a controller is capable of satisfying a desired system requirement, specified using Signal Temporal Logic (STL), under possible deviations in the system. Based on this definition, we propose a novel analysis problem, called the tolerance falsification problem, which involves finding small deviations that result in a violation of the given requirement. We present a novel, two-layer simulation-based analysis framework and a novel search heuristic for finding small tolerance violations. To evaluate our approach, we construct a set of benchmark problems where system parameters can be configured to represent different types of uncertainties and disturbancesin the system. Our evaluation shows that our falsification approach and heuristic can effectively find small tolerance violations.

Pedro Mendes, Maria Casimiro, Paolo Romano et al.

In the literature on hyper-parameter tuning, a number of recent solutions rely on low-fidelity observations (e.g., training with sub-sampled datasets) in order to efficiently identify promising configurations to be then tested via high-fidelity observations (e.g., using the full dataset). Among these, HyperBand is arguably one of the most popular solutions, due to its efficiency and theoretically provable robustness. In this work, we introduce HyperJump, a new approach that builds on HyperBand's robust search strategy and complements it with novel model-based risk analysis techniques that accelerate the search by skipping the evaluation of low risk configurations, i.e., configurations that are likely to be eventually discarded by HyperBand. We evaluate HyperJump on a suite of hyper-parameter optimization problems and show that it provides over one-order of magnitude speed-ups, both in sequential and parallel deployments, on a variety of deep-learning, kernel-based learning, and neural architectural search problems when compared to HyperBand and to several state-of-the-art optimizers.

Rebekka Wohlrab, David Garlan

[Context and motivation:] For realistic self-adaptive systems, multiple quality attributes need to be considered and traded off against each other. These quality attributes are commonly encoded in a utility function, for instance, a weighted sum of relevant objectives. [Question/problem:] The research agenda for requirements engineering for self-adaptive systems has raised the need for decision-making techniques that consider the trade-offs and priorities of multiple objectives. Human stakeholders need to be engaged in the decision-making process so that the relative importance of each objective can be correctly elicited. [Principal ideas/results:] This research preview paper presents a method that supports multiple stakeholders in prioritizing relevant quality attributes, negotiating priorities to reach an agreement, and giving input to define utility functions for self-adaptive systems. [Contribution:] The proposed method constitutes a lightweight solution for utility function definition. It can be applied by practitioners and researchers who aim to develop self-adaptive systems that meet stakeholders' requirements. We present details of our plan to study the application of our method using a case study.

Pedro Mendes, Maria Casimiro, Paolo Romano et al.

This work introduces TrimTuner, the first system for optimizing machine learning jobs in the cloud to exploit sub-sampling techniques to reduce the cost of the optimization process while keeping into account user-specified constraints. TrimTuner jointly optimizes the cloud and application-specific parameters and, unlike state of the art works for cloud optimization, eschews the need to train the model with the full training set every time a new configuration is sampled. Indeed, by leveraging sub-sampling techniques and data-sets that are up to 60x smaller than the original one, we show that TrimTuner can reduce the cost of the optimization process by up to 50x. Further, TrimTuner speeds-up the recommendation process by 65x with respect to state of the art techniques for hyper-parameter optimization that use sub-sampling techniques. The reasons for this improvement are twofold: i) a novel domain specific heuristic that reduces the number of configurations for which the acquisition function has to be evaluated; ii) the adoption of an ensemble of decision trees that enables boosting the speed of the recommendation process by one additional order of magnitude.

Roykrong Sukkerd, Reid Simmons, David Garlan

End-users' trust in automated agents is important as automated decision-making and planning is increasingly used in many aspects of people's lives. In real-world applications of planning, multiple optimization objectives are often involved. Thus, planning agents' decisions can involve complex tradeoffs among competing objectives. It can be difficult for the end-users to understand why an agent decides on a particular planning solution on the basis of its objective values. As a result, the users may not know whether the agent is making the right decisions, and may lack trust in it. In this work, we contribute an approach, based on contrastive explanation, that enables a multi-objective MDP planning agent to explain its decisions in a way that communicates its tradeoff rationale in terms of the domain-level concepts. We conduct a human subjects experiment to evaluate the effectiveness of our explanation approach in a mobile robot navigation domain. The results show that our approach significantly improves the users' understanding, and confidence in their understanding, of the tradeoff rationale of the planning agent.

Javier Cámara, Alessandro V. Papadopoulos, Thomas Vogel et al.

Two of the main paradigms used to build adaptive software employ different types of properties to capture relevant aspects of the system's run-time behavior. On the one hand, control systems consider properties that concern static aspects like stability, as well as dynamic properties that capture the transient evolution of variables such as settling time. On the other hand, self-adaptive systems consider mostly non-functional properties that capture concerns such as performance, reliability, and cost. In general, it is not easy to reconcile these two types of properties or identify under which conditions they constitute a good fit to provide run-time guarantees. There is a need of identifying the key properties in the areas of control and self-adaptation, as well as of characterizing and mapping them to better understand how they relate and possibly complement each other. In this paper, we take a first step to tackle this problem by: (1) identifying a set of key properties in control theory, (2) illustrating the formalization of some of these properties employing temporal logic languages commonly used to engineer self-adaptive software systems, and (3) illustrating how to map key properties that characterize self-adaptive software systems into control properties, leveraging their formalization in temporal logics. We illustrate the different steps of the mapping on an exemplar case in the cloud computing domain and conclude with identifying open challenges in the area.

Nabor C. Mendonca, Pooyan Jamshidi, David Garlan et al.

A self-adaptive system can dynamically monitor and adapt its behavior to preserve or enhance its quality attributes under uncertain operating conditions. This article identifies key challenges for the development of microservice applications as self-adaptive systems, using a cloud-based intelligent video surveillance application as a motivating example. It also suggests potential new directions for addressing most of the identified challenges by leveraging existing microservice practices and technologies.

Pooyan Jamshidi, Javier Cámara, Bradley Schmerl et al.

Modern cyber-physical systems (e.g., robotics systems) are typically composed of physical and software components, the characteristics of which are likely to change over time. Assumptions about parts of the system made at design time may not hold at run time, especially when a system is deployed for long periods (e.g., over decades). Self-adaptation is designed to find reconfigurations of systems to handle such run-time inconsistencies. Planners can be used to find and enact optimal reconfigurations in such an evolving context. However, for systems that are highly configurable, such planning becomes intractable due to the size of the adaptation space. To overcome this challenge, in this paper we explore an approach that (a) uses machine learning to find Pareto-optimal configurations without needing to explore every configuration and (b) restricts the search space to such configurations to make planning tractable. We explore this in the context of robot missions that need to consider task timeliness and energy consumption. An independent evaluation shows that our approach results in high-quality adaptation plans in uncertain and adversarial environments.

Marco Autili, Paola Inverardi, Massimo Tivoli et al.

Adaptation of software components is an important issue in Component Based Software Engineering (CBSE). Building a system from reusable or Commercial-Off-The-Shelf (COTS) components introduces a set of problems, mainly related to compatibility and communication aspects. On one hand, components may have incompatible interaction behavior. This might require to restrict the system's behavior to a subset of safe behaviors. On the other hand, it might be necessary to enhance the current communication protocol. This might require to augment the system's behavior to introduce more sophisticated interactions among components. We address these problems by enhancing our architectural approach which allows for detection and recovery of incompatible interactions by synthesizing a suitable coordinator. Taking into account the specification of the system to be assembled and the specification of the protocol enhancements, our tool (called SYNTHESIS) automatically derives, in a compositional way, the glue code for the set of components. The synthesized glue code implements a software coordinator which avoids incompatible interactions and provides a protocol-enhanced version of the composed system. By using an assume-guarantee technique, we are able to check, in a compositional way, if the protocol enhancement is consistent with respect to the restrictions applied to assure the specified safe behaviors.

David Garlan, Vishal Dwivedi, Ivan Ruchkin et al.

Within an increasing number of domains an important emerging need is the ability for technically naive users to compose computational elements into novel configurations. Examples include astronomers who create new analysis pipelines to process telescopic data, intelligence analysts who must process diverse sources of unstructured text to discover socio-technical trends, and medical researchers who have to process brain image data in new ways to understand disease pathways. Creating such compositions today typically requires low-level technical expertise, limiting the use of computational methods and increasing the cost of using them. In this paper we describe an approach - which we term end-user architecting - that exploits the similarity between such compositional activities and those of software architects. Drawing on the rich heritage of software architecture languages, methods, and tools, we show how those techniques can be adapted to support end users in composing rich computational systems through domain-specific compositional paradigms and component repositories, without requiring that they have knowledge of the low-level implementation details of the components or the compositional infrastructure. Further, we outline a set of open research challenges that the area of end-user architecting raises.