United States

As AI systems take on critical roles in high-stakes decisions, the ability to explain their decisions becomes crucial for assigning appropriate trust. Limitations for explaining black-box systems make interpretable AI methods an appealing alternative. Case-based reasoning (CBR) solves new problems by retrieving solutions to similar prior problems and adapting their solutions to fit the new problem circumstances, and is “naturally interpretable”, in that decisions can be justified by presenting the cases from which they were derived. However, there has been little study of when the benefits of CBR apply and how specifics of case presentation affect explanatory effectiveness. To assess factors affecting the usefulness of cases to explain decision-making, we conducted a human subjects study (N=110) to investigate how alternative types of explanations of a CBR system’s conclusions impact user trust in the CBR system. Our goal was to inform future design of case-based decision support systems by advancing understanding of how different system types and explanations affect the factors of goodness, convincingness and trust in the system.

Our experiment tested how two different types of interactions with a CBR system—one-shot, or interactive “conversational case-based reasoning”—and four types of explanation presentation affected subjects’ assessment of explanations of decisions of a case-based classification system. In our tests, augmenting presented cases with explicitly highlighted similarities and differences was helpful, but augmenting cases with nearby counterfactual cases was not. Whether the system interacted in a conversational manner was not found to impact scores, suggesting that the case itself is more important to users than how it was retrieved. To assess effects when erroneous conclusions were explained, we provided the user with an intentionally incorrect conclusion, without the user being made aware that the conclusion was incorrect, and found impact on scores for goodness, convincingness and trust of case-based explanations. This suggested that case-based explanations were not masking system errors. We also studied the effects of similarity of the presented case to the new problem, and found that it impacted convincingness and trust, underlining the importance of sufficient past cases.

The study illustrates that even for interpretable methods, specifics of explanation presentation matter. Results can be used by system designers to improve the quality of the system explanations. Also, by underlining that the key factor in goodness, convincingness, and trust is the quality of the cases themselves, the study provides support for case presentation as a natural vehicle for explanation.

This work was funded by the US Department of Defense (Contract W52P1J2093009), and by the Department of the Navy, Office of Naval Research (Award N00014-19-1-2655).


NAML 2024

Interpretable Decision-Making By Case-Based Reasoning: Lessons from a Human Subjects Study

human subjects study

case-based reasoning

explainable ai

technical paper

### Welcome to the 8th annual workshop on Naval Applications of Machine Learning!

NAML is the premier annual event showcasing current machine learning research relevant to Naval applications. We welcome participation from the defense community, industry, and academia. The NAML workshop is an opportunity for researchers from Navy research and development centers and our partner organizations to exchange technical ideas and to come together as a team.

The goals of NAML are:

* To spread awareness of current machine learning research relevant to Naval applications

* To connect machine learning researchers with experts in military requirements

* To build and strengthen collaborations within the defense research community

Naval Information Warfare Center Pacific is partnering with AFCEA San Diego chapter to host the workshop.

**General Chairs:**

Katie Rainey, Naval Information Warfare Center Pacific

Nicole Isoda, Naval Information Warfare Center Pacific 

**[More about NAML 2024](https://sites.google.com/go.spawar.navy.mil/naml/)**

**NEW! Join 700+ people - Receive a Special Discount **

Now you can have access to all the lectures & posters and even engage with the speakers (send them emails, ask questions) on the Underline platform. 
Access to NAML 2024 was originally $599 in person. We are offering a one-time only discounted on-demand access to this NAML 2024 content **for only $99** (complimentary to students, government, and military). 
 
**To register, please click** [**here**](https://naml2024.paperform.co/)

The 8th annual workshop on Naval Applications of Machine Learning is an opportunity for researchers from Navy research and development centers and our partner organizations to exchange technical ideas and to come together as a team. The goals of NAML are to spread awareness of current machine learning research relevant to Naval applications, to connect machine learning researchers with experts in military requirements and to build and strengthen collaborations within the defense research community.


"Artificial intelligence (AI) and machine learning (ML) have grown in an unprecedented manner, leading to their widespread adoption in complex domain settings far removed from computer science and computer engineering. As these technologies have matured interest has increased in the development of artificial cognitive capabilities that would allow these new systems to act less like an automated appliance, and more like an interdependent teammate. Similarly, advances in team science and cognitive science have led to advances in our understanding of the patterns and frameworks human teammates utilize in communication, coordination, and adaptability to form effective teams even with diverse team composition, and novel task environments. While much of the literature on effective teaming has focused on the realm of human teaming, these advances have been noted as important to human-machine teams as well.
 
In order to be effective as interdependent teammates, instead of merely statistical appliances, machine co-performers need to develop skills for teaming intelligence including the establishment of common ground with human co-performers and understanding their purpose in joint activity (Clark, 1996).  In response to these needs, we propose a novel framework that integrates dynamic Self-Organizing Maps (SOMs) and the automated generation and extraction of team relevant ontologies to generate common ground and teaming intelligence aligned to the intent of human commanders. This framework enhances AI capabilities by providing contextually rich and explainable insights, fostering a shared understanding and seamless interaction between human operators and AI systems.  We demonstrate this framework using over open source intelligence data collected from gray zone conflicts in the domain of  Illegal, Unreported, and Unregulated (IUU) fishing activity, including 55 billion AIS messages representing over 200,000 industrial fishing vessels, carriers, and bunker vessels from 2012 - 2017 with spatiotemporal characteristics including over 100 billion GPS positions providing full labeled tracks of voyages.  "

Enhancing Explainability for Human-Machine Teams with Automatic Common Ground Establishment and Ontology Estimation

The time- and computational-cost of solving the atmospheric radiative transfer problem with sufficient spatial and temporal resolution for predictive-quality scene simulation has been a limiting factor for many decades. 
The advent of deep learning has shown its ability to dramatically accelerate the inference times, sometimes running 100s of times faster, while still achieving extremely high accuracy compared to traditional reference/standard numerical solvers. 
We will present unprecedented preliminary results of a novel deep-learning model architecture for hyperspectral atmospheric radiative transfer solution estimation along with a quantitative analysis of its performance, both in terms of accuracy/precision and runtime

Preliminary results for a Deep Learning Atmospheric Radiative Transfer Solver to accelerate Predictive Scene Simulation

Point measurements of optical turbulence rely on sensitive instruments that are often difficult to deploy in many environments. Furthermore, existing models of turbulence may not perform well in those environments. In a recent experiment at the Naval Postgraduate School, optical turbulence measurements using a sonic anemometer along with various atmospheric parameters (air temperature, humidity, solar flux, etc.) were collected over a ten-month period. 
Using the observed atmospheric parameters as predictors, we developed a bagged ensemble of regression trees using optical turbulence as the response. Feature selection was used to decide the parameter set. Overall, the model’s predicted values showed good agreement with observed values of turbulence derived from the sonic anemometer data.


Prediction of Optical Turbulence Using Machine Learning Regression Models

Atmospheric propagation effects, including optical turbulence, impact performance of Naval systems including optical communications, laser, and imaging systems. Cn2 is a common quantitative measure used to characterize optical turbulence and its impact on optical systems. Parametric models of turbulence (such as NAVSLaM) generally perform well under well-characterized, relatively homogeneous environments, however are not typically designed for challenging environments, with varying terrain height. Machine learning is well suited to modeling these kinds of scenarios.

In this presentation, we will demonstrate Cn2 predictions along an inhomogeneous test path, by building and applying specific ML (machine learning) models, using meteorological data as input parameters. We use approximately two years of data collected on the SPARTA test path in San Diego, CA, a path that is well instrumented with meteorological and turbulence sensors. The extensive dataset provides a large training set, and enables diurnal and seasonal trend analysis. We explore different combinations of inputs and rank them in order of their contribution and relative importance to the model prediction. A comparison between various ML algorithms, such as LightGBM (Light Gradient Boosting Machine) and Extra Tree, is performed to determine which best predicts Cn2. In addition to using meteorological data as inputs to the model, we also experiment with using images captured at a test range setup at NIWC Pacific in San Diego.

Novel Approaches to Modeling Turbulence in Inhomogeneous Environments

Atmospheric turbulence is a phenomenon resulting from temperature, distance, altitude, humidity, wind speed, pollution, and fog, which creates aberrations in laser optical systems that attempt to track a target. With atmospheric turbulences, the ability to focus radiated energy reliably and repeatedly at range, with precision and controllable effects, is hampered due to the aberrations created for long range telescopic systems. Current approaches to correct for aberrations commonly utilize a beacon projected onto the target. However, this approach is undesirable for deployed systems, since using a beacon reveals the location of the detector. The objective of this work is to develop a beaconless, imaging-based method of high-bandwidth wavefront sensor that corrects for atmosphere-induced aberrations in long range telescopic systems.

The breakdown of the workflow is as follows. In order to understand the atmospheric turbulence effects on long range telescopic systems, the first step entails creating the aberrations in images with Zernike polynomials, which are functions that create various distortions in aperture lenses. The approach includes synthetic generation of the dataset in MATLAB that varies the degree of phase and tilt, two controllable parameters that create the varying distortions. The second part investigates image compression techniques for storing sparse representations of the images. The objectives for image compression is to discover a representation that retains the needed information to perform the image correction while utilizing less memory storage and less computational time to process, since this correction will need to be performed in a real-time adaptive optics system. The third part involves performing the image deblurring of the aberrations. This step requires performing a deconvolution of the image by separating the image from the point spread function that induces the aberrations, and performing image restoration that removes the aberrations. This work presents results that ties in this closed-loop workflow for an adaptive optics system by comparing several image deblurring and compression approaches, accounting for execution speed and accuracy of the resulting image.

Image Based Wavefront Sensor for Atmospheric Turbulences

Truly autonomous operation of unmanned underwater vehicles (UUVs) is of critical importance as the Navy seeks to implement its 30-year Distributed Maritime Operations strategy. During the course of a mission, unmanned systems face the need to make decisions that range from when, where, and how to position themselves and operate their mission systems in the face of uncertain, unknown, and/or changing environmental conditions. Many in-service systems, however, lack a well-established means of incorporating in-situ observations of environmental variables, such as local or global currents, into their path planning, guidance, and/or control decisions. As such, there is a need for technologies that can autonomously make environmentally aware decisions for the UUV in order to improve the probability of mission completion.

We describe here a deep reinforcement learning (RL) framework that partially addresses these issues by generating energy-minimizing guidance systems for UUVs. Specifically, the proposed framework outputs trained policy networks (agents) capable of mapping local current velocity measurements to heading, depth, and speed commands that yield nearly-optimal, minimum-energy transits through spatially varying current fields. Numerical simulations in two and three spatial dimensions demonstrate that the energy cost of transits executed under the guidance of a trained agent are better than dead-reckoning, a common in-service guidance scheme, and approach known optimal performance for a set of realistic test scenarios that were not included in the training process.


A Novel Reinforcement Learning Framework for Improved Guidance and Power Management of Unmanned Underwater Vehicles

"The Intelligence Advanced Research Projects Activity (IARPA) SCISRS program addresses the critical need for detecting and characterizing RF anomalies indicative of secure data compromise in complex RF environments. The program focuses on identifying low probability of intercept (LPI) signals, altered or mimicked signals, and unintentional emissions. Expedition Technology (EXP) supports this initiative with our solution Sensing Hardware Emissions via Artificially Intelligent RF Reasoning Systems (SHEARRS). SHEARRS integrates traditional signal processing and machine learning in an end-to-end system to enhance RF security. This work builds off the foundational ML capabilities developed under the DARPA RFMLS program and ongoing transition sponsored by Naval Information Warfare Center (NIWC) Pacific. 

SHEARRS is designed to be scalable, computationally efficient, and adaptable to diverse radio hardware. The system operates by applying advanced algorithms to detect subtle, anomalous signals amidst a noisy background across a large frequency range of interest (100 MHz to 6 GHz) using commodity COTS software defined radios (SDR). It has been tested and validated on government RF testbeds, demonstrating its ability to identify LPI signals, recognize altered transmissions, and detect unintentional emissions from electronic devices. 

Our proposed presentation for the Eighth Annual Workshop on Naval Applications of Machine Learning will detail the technical architecture of SHEARRS, its operational workflow, and the methodologies employed for anomaly detection and characterization. This will include discussions of our modulation recognition neural network capabilities for narrow and wideband scenes, statistical signal processing for efficient detection of spectral anomalies using a fast sweeping 6 GHz spectrum monitor, and our end-to-end system pipeline for combining the outputs of multiple detection engines. We will discuss the results from the IARPA testbed evaluations, emphasizing the system's performance metrics, scalability, and adaptability. 

The integration of SHEARRS with the SCISRS program represents a significant advancement in securing data in complex RF environments. Our solution not only detects but also characterizes a wide spectrum of RF anomalies, providing a robust tool for national security missions operating in diverse and uncontrollable environments. The presentation aims to inform on the technical capabilities, applications, and future developments of SHEARRS within the context of the SCISRS program. "

Sensing Hardware Emissions and Anomalies via Artificially Intelligent RF Reasoning Systems (SHEAARRS)

NPS team developed a vessel tracking system that uses AIS, satelite imagery, and electromagnetics to monitor traffic in waterways, ports, and coastal areas. Working with open-source and vendor data, a common operating picture was provided that prevents AIS spoofing and also allows classification based on AIS and vessel electromagnetics.

Common Operating Picture of Maritime Vessels using AIS, Satellite Imagery and Electro Magnetics

A novel approach for pulse-Doppler radar moving target detection in range-Doppler maps is proposed. The proposed ensemble architecture is highly modular, generalizable, and extensible – allowing for customization and applicability to a wide variety of radar systems. It contains a suite of statistical, convolutional, and graph-based machine learning models which seamlessly combine to detect targets in real-world range-Doppler maps. The model has been shown to be adjustable and tunable to work on multiple real-world datasets, including those containing air-to-air targets and air-to-ground targets, and has a high probability of detection and low false alarm rate in all cases.

The RAnge-Doppler Detection (RADD) Model

This paper presents an eXplainable Deep Reinforcement Learning (XDRL) based strategy for solving the proposed problem of fleet-level aircraft maintenance scheduling (AMS) optimization. The XDRL-AMS considers various factors such as the aircraft’s initial status, mission requirements, maintenance resource capacity, and operational constraints to create a maintenance schedule for a specified period. The schedule aims to balance both mission readiness and cost reduction. We developed an RL environment, called AMS-Gym, using the OpenAI Gym toolkit specifically designed for this problem. AMS-Gym is highly flexible, allowing for easy extension to more complex scenarios and incorporating additional explanatory capabilities. The explainable RL capability was achieved by utilizing a decomposed reward Deep Q-Network (drDQN) algorithm. In the context of the AMS scenario, the drDQN consists of two parts: (i) a DQN that aims to maximize the mission accomplishment objective, and (ii) a DQN that aims to minimize the maintenance cost objective. As a result, the proposed drDQN strategy can generate real-time aircraft maintenance decisions, explain why those decisions were selected, and present the tradeoffs between the chosen action and non-selected alternatives. Experiment results show that the proposed drDQN performs well, providing an approximate solution to the vanilla DQN with a simpler structure while offering the ability to explain its decisions. In addition, a web-based prototype with an intuitive textual and visual user interface was developed to demonstrate the feasibility of the drDQN approach.

Premium content

Interpretable Decision-Making By Case-Based Reasoning: Lessons from a Human Subjects Study

Downloads

Next from NAML 2024

Enhancing Explainability for Human-Machine Teams with Automatic Common Ground Establishment and Ontology Estimation

Stay up to date with the latest Underline news!

PRESENTATIONS

CONFERENCES

COMPANY

RESOURCES