Matching Items (43)
Description
Rapid advancements in artificial intelligence (AI) have revolutionized various do- mains, enabling the development of sophisticated models capable of solving complex problems. However, as AI systems increasingly participate in critical decision-making processes, concerns about their interpretability, robustness, and reliability have in- tensified. Interpretable AI models, such as the Concept-Centric Transformer (CCT), have emerged as promising solutions to enhance transparency in AI models. Yet, in- creasing model interpretability often requires enriching training data with concept ex- planations, escalating training costs. Therefore, intrinsically interpretable models like CCT must be designed to be data-efficient, generalizable—to accommodate smaller training sets—and robust against noise and adversarial attacks. Despite progress in interpretable AI, ensuring the robustness of these models remains a challenge.This thesis enhances the data efficiency and generalizability of the CCT model by integrating four techniques: Perturbation Random Masking (PRM), Attention Random Dropout (ARD), and the integration of manifold mixup and input mixup for memory broadcast. Comprehensive experiments on benchmark datasets such as CIFAR-100, CUB-200-2011, and ImageNet show that the enhanced CCT model achieves modest performance improvements over the original model when using a full training set. Furthermore, this performance gap increases as the training data volume decreases, particularly in few-shot learning scenarios. The enhanced CCT maintains high accuracy with limited data (even without explicitly training on ex- ample concept-level explanations), demonstrating its potential for real-world appli- cations where labeled data are scarce. These findings suggest that the enhancements enable more effective use of CCT in settings with data constraints. Ablation studies reveal that no single technique—PRM, ARD, or mixups—dominates in enhancing performance and data efficiency. Each contributes nearly equally, and their combined application yields the best results, indicating a synergistic effect that bolsters the model’s capabilities without any single method being predominant.
The results of this research highlight the efficacy of the proposed enhancements in refining CCT models for greater performance, robustness, and data efficiency. By demonstrating improved performance and resilience, particularly in data-limited sce- narios, this thesis underscores the practical applicability of advanced AI systems in critical decision-making roles.
ContributorsPark, Keun Hee (Author) / Pavlic, Theodore (Thesis advisor) / Choi, YooJung (Committee member) / Yang, Yezhou (Committee member) / Arizona State University (Publisher)
Created2024
Description
Robot motion and control remains a complex problem both in general and inthe field of machine learning (ML). Without ML approaches, robot controllers are
typically designed manually, which can take considerable time, generally requiring
accounting for a range of edge cases and often producing models highly constrained
to specific tasks. ML can decrease the time it takes to create a model while simultaneously allowing it to operate on a broader range of tasks. The utilization of neural
networks to learn from demonstration is, in particular, an approach with growing
popularity due to its potential to quickly fit the parameters of a model to mimic
training data.
Many such neural networks, especially in the realm of transformer-based architectures, act more as planners, taking in an initial context and then generating a
sequence from that context one step at a time. Others hybridize the approach, predicting a latent plan and conditioning immediate actions on that plan. Such approaches may limit a model’s ability to interact with a dynamic environment, needing to replan to fully update its understanding of the environmental context. In this
thesis, Language-commanded Scene-aware Action Response (LanSAR) is proposed as
a reactive transformer-based neural network that makes immediate decisions based
on previous actions and environmental changes. Its actions are further conditioned
on a language command, serving as a control mechanism while also narrowing the
distribution of possible actions around this command. It is shown that LanSAR successfully learns a strong representation of multimodal visual and spatial input, and
learns reasonable motions in relation to most language commands. It is also shown
that LanSAR can struggle with both the accuracy of motions and understanding the
specific semantics of language commands
ContributorsHardy, Adam (Author) / Ben Amor, Heni (Thesis advisor) / Srivastava, Siddharth (Committee member) / Pavlic, Theodore (Committee member) / Arizona State University (Publisher)
Created2024
Description
Social insects collectively exploit food sources by recruiting nestmates, creating positive feedback that steers foraging effort to the best locations. The nature of this positive feedback varies among species, with implications for collective foraging. The mass recruitment trails of many ants are nonlinear, meaning that small increases in recruitment effort yield disproportionately large increases in recruitment success. The waggle dance of honeybees, in contrast, is believed to be linear, meaning that success increases proportionately to effort. However, the implications of this presumed linearityhave never been tested. One such implication is the prediction that linear recruiters will equally exploit two identical food sources, in contrast to nonlinear recruiters, who randomly choose only one of them. I tested this prediction in colonies of honeybees that were isolated in flight cages and presented with two identical sucrose feeders. The results from 15 trials were consistent with linearity, with many cases of equal exploitation of the feeders. In addition, I tested the prediction that linear recruiters can reallocate their forager distribution when unequal feeders are swapped in position. Results from 15 trials were consistent with linearity, with many cases of clear choice for a stronger food source, followed by a subsequent switch with reallocation of foragers to the new location of the stronger food source. These findings show evidence of a linear pattern of nestmate recruitment, with implications for how colonies effectively distribute their foragers across
available resources.
ContributorsAlam, Showmik (Author) / Shaffer, Zachary (Thesis advisor) / Pratt, Stephen C (Thesis advisor) / Ozturk, Cahit (Committee member) / Pavlic, Theodore (Committee member) / Arizona State University (Publisher)
Created2022
Description
In recent years, the development of Control Barrier Functions (CBF) has allowed safety guarantees to be placed on nonlinear control affine systems. While powerful as a mathematical tool, CBF implementations on systems with high relative degree constraints can become too computationally intensive for real-time control. Such deployments typically rely on the analysis of a system's symbolic equations of motion, leading to large, platform-specific control programs that do not generalize well. To address this, a more generalized framework is needed. This thesis provides a formulation for second-order CBFs for rigid open kinematic chains. An algorithm for numerically computing the safe control input of a CBF is then introduced based on this formulation. It is shown that this algorithm can be used on a broad category of systems, with specific examples shown for convoy platooning, drone obstacle avoidance, and robotic arms with large degrees of freedom. These examples show up to three-times performance improvements in computation time as well as 2-3 orders of magnitude in the reduction in program size.
ContributorsPietz, Daniel Johannes (Author) / Fainekos, Georgios (Thesis advisor) / Vrudhula, Sarma (Thesis advisor) / Pedrielli, Giulia (Committee member) / Pavlic, Theodore (Committee member) / Arizona State University (Publisher)
Created2022
Description
Aggregation is a fundamental principle of animal behavior; it is especially significant tohighly social species, like ants. Ants typically aggregate their workers and brood in a central nest, potentially due to advantages in colony defense and regulation of the environment. In many ant species, when a colony must abandon its nest, it can effectively reach consensus on a new home. Ants of the genus Temnothorax have become a model for this collective decision-making process, and for decentralized cognition more broadly. Previous studies examine emigration by well-aggregated colonies, but can these ants also reach consensus when the colony has been scattered? Such scattering may readily occur in nature if the nest is disturbed by natural or man- made disasters. In this exploratory study, Temnothorax rugatulus colonies were randomly scattered in an arena and presented with a binary equal choice of nest sites. Findings concluded that the colonies were able to re-coalesce, however consensus is more difficult than for aggregated colonies and involved an additional primary phase of multiple temporary aggregations eventually yielding to reunification. The maximum percent of colony utilization for these aggregates was reached within the first hour, after which point, consensus tended to rise as aggregation decreased. Small, but frequent, aggregates formed within the first twenty minutes and remained and dissolved to the nest by varying processes. Each colony included a clump containing the queen, with the majority of aggregates containing at least one brood item. These findings provide additional insight to house-hunting experiments in more naturally
challenging circumstances, as well as aggregation within Temnothorax colonies.
ContributorsGoodland, Brooke (Author) / Shaffer, Zachary (Thesis advisor) / Pratt, Stephen (Thesis advisor) / Pavlic, Theodore (Committee member) / Arizona State University (Publisher)
Created2023
Description
A Graph Neural Network (GNN) is a type of neural network architecture that operates on data consisting of objects and their relationships, which are represented by a graph. Within the graph, nodes represent objects and edges represent associations between those objects. The representation of relationships and correlations between data is unique to graph structures. GNNs exploit this feature of graphs by augmenting both forms of data, individual and relational, and have been designed to allow for communication and sharing of data within each neural network layer. These benefits allow each node to have an enriched perspective, or a better understanding, of its neighbouring nodes and its connections to those nodes. The ability of GNNs to efficiently process high-dimensional node data and multi-faceted relationships among nodes gives them advantages over neural network architectures such as Convolutional Neural Networks (CNNs) that do not implicitly handle relational data. These quintessential characteristics of GNN models make them suitable for solving problems in which the correspondences among input data are needed to produce an accurate and precise representation of these data. GNN frameworks may significantly improve existing communication and control techniques for multi-agent tasks by implicitly representing not only information associated with the individual agents, such as agent position, velocity, and camera data, but also their relationships with one another, such as distances between the agents and their ability to communicate with one another. One such task is a multi-agent navigation problem in which the agents must coordinate with one another in a decentralized manner, using proximity sensors only, to navigate safely to their intended goal positions in the environment without collisions or deadlocks. The contribution of this thesis is the design of an end-to-end decentralized control scheme for multi-agent navigation that utilizes GNNs to prevent inter-agent collisions and deadlocks. The contributions consist of the development, simulation and evaluation of the performance of an advantage actor-critic (A2C) reinforcement learning algorithm that employs actor and critic networks for training that simultaneously approximate the policy function and value function, respectively. These networks are implemented using GNN frameworks for navigation by groups of 3, 5, 10 and 15 agents in simulated two-dimensional environments. It is observed that in $40\%$ to $50\%$ of the simulation trials, between 70$\%$ to 80$\%$ of the agents reach their goal positions without colliding with other agents or becoming trapped in deadlocks. The model is also compared to a random run simulation, where actions are chosen randomly for the agents and observe that the model performs notably well for smaller groups of agents.
ContributorsAyalasomayajula, Manaswini (Author) / Berman, Spring (Thesis advisor) / Mian, Sami (Committee member) / Pavlic, Theodore (Committee member) / Arizona State University (Publisher)
Created2022
Description
In certain ant species, groups of ants work together to transport food and materials back to their nests. In some cases, the group exhibits a leader-follower behavior in which a single ant guides the entire group based on its knowledge of the destination. In some cases, the leader role is occupied temporarily by an ant, only to be replaced when an ant with new information arrives. This kind of behavior can be very useful in uncertain environments where robot teams work together to transport a heavy or bulky payload. The purpose of this research was to study ways to implement this behavior on robot teams.
In this work, I combined existing dynamical models of collective transport in ants to create a stochastic model that describes these behaviors and can be used to control multi-robot systems to perform collective transport. In this model, each agent transitions stochastically between roles based on the force that it senses the other agents are applying to the load. The agent’s motion is governed by a proportional controller that updates its applied force based on the load velocity. I developed agent-based simulations of this model in NetLogo and explored leader-follower scenarios in which agents receive information about the transport destination by a newly informed agent (leader) joining the team. From these simulations, I derived the mean allocations of agents between “puller” and “lifter” roles and the mean forces applied by the agents throughout the motion.
From the simulation results obtained, we show that the mean ratio of lifter to puller populations is approximately 1:1. We also show that agents using the role update procedure based on forces are required to exert less force than agents that select their role based on their position on the load, although both strategies achieve similar transport speeds.
In this work, I combined existing dynamical models of collective transport in ants to create a stochastic model that describes these behaviors and can be used to control multi-robot systems to perform collective transport. In this model, each agent transitions stochastically between roles based on the force that it senses the other agents are applying to the load. The agent’s motion is governed by a proportional controller that updates its applied force based on the load velocity. I developed agent-based simulations of this model in NetLogo and explored leader-follower scenarios in which agents receive information about the transport destination by a newly informed agent (leader) joining the team. From these simulations, I derived the mean allocations of agents between “puller” and “lifter” roles and the mean forces applied by the agents throughout the motion.
From the simulation results obtained, we show that the mean ratio of lifter to puller populations is approximately 1:1. We also show that agents using the role update procedure based on forces are required to exert less force than agents that select their role based on their position on the load, although both strategies achieve similar transport speeds.
ContributorsGah, Elikplim (Author) / Berman, Spring M (Thesis advisor, Committee member) / Pavlic, Theodore (Committee member) / Marvi, Hamidreza (Committee member) / Arizona State University (Publisher)
Created2020
Description
The desert ant, Novomessor albisetosus, is an ideal model system for studying collective transport in ants and self-organized cooperation in natural systems. Small teams collect and stabilize around objects encountered by these colonies in the field, and the teams carry them in straight paths at a regulated velocity back to nearby nest entrances. The puzzling finding that teams are slower than individuals contrasts other cases of cooperative transport in ants. The statistical distribution of speeds has been found to be consistent with the slowest-ant model, but the key assumption that individual ants consistently vary in speed has not been tested. To test this, information is needed about the natural distribution of individual ant speeds in colonies and whether some ants are intrinsically slow or fast. To investigate the natural, individual-level variation in ants carrying loads, data were collected on single workers carrying fig seeds in arenas separated from other workers. Using three separate, small arenas, the instantaneous speed of each seed-laden worker was recorded when she picked up a fig seed and transported within the arena. Instantaneous speeds were measured by dividing the distance traveled in each frame by how much time had passed.
There were nine ants who transported a fig seed numerous times and there was a clear variation in their average instantaneous speed. Within an ant, slightly varying speeds were found as well, but within-ant speeds were not as varied as speed across ants. These results support the conclusion that there is intrinsic variation in the speed of an individual which supports the slowest-ant model, but this may require further experimentation to test thoroughly. This information aids in the understanding of the natural variation of ants cooperatively carrying larger loads in groups.
There were nine ants who transported a fig seed numerous times and there was a clear variation in their average instantaneous speed. Within an ant, slightly varying speeds were found as well, but within-ant speeds were not as varied as speed across ants. These results support the conclusion that there is intrinsic variation in the speed of an individual which supports the slowest-ant model, but this may require further experimentation to test thoroughly. This information aids in the understanding of the natural variation of ants cooperatively carrying larger loads in groups.
ContributorsCastro, Samantha (Author) / Pavlic, Theodore (Thesis director) / Pratt, Stephen (Committee member) / School of Life Sciences (Contributor) / Barrett, The Honors College (Contributor)
Created2020-12
Description
Ultimate Frisbee or "Ultimate," is a fast growing field sport that is being played competitively at universities across the country. Many mid-tier college teams have the goal of winning as many games as possible, however they also need to grow their program by training and retaining new players. The purpose of this project was to create a prototype statistical tool that maximizes a player line-up's probability of scoring the next point, while having as equal playing time across all experienced and novice players as possible. Game, player, and team data was collected for 25 different games played over the course of 4 tournaments during Fall 2017 and early Spring 2018 using the UltiAnalytics iPad application. "Amount of Top 1/3 Players" was the measure of equal playing time, and "Line Efficiency" and "Line Interaction" represented a line's probability of scoring. After running a logistic regression, Line Efficiency was found to be the more accurate predictor of scoring outcome than Line Interaction. An "Equal PT Measure vs. Line Efficiency" graph was then created and the plot showed what the optimal lines were depending on what the user's preferences were at that point in time. Possible next steps include testing the model and refining it as needed.
ContributorsSpence, Andrea Nicole (Author) / McCarville, Daniel R. (Thesis director) / Pavlic, Theodore (Committee member) / Industrial, Systems and Operations Engineering Program (Contributor) / Barrett, The Honors College (Contributor)
Created2018-05
Description
Every year, millions of guests visit theme parks internationally. Within that massive population, accidents and emergencies are bound to occur. Choosing the correct location for emergency responders inside of the park could mean the difference between life and death. In an effort to provide the utmost safety for the guests of a park, it is important to make the best decision when selecting the location for emergency response crews. A theme park is different from a regular residential or commercial area because the crowds and shows block certain routes, and they change throughout the day. We propose an optimization model that selects staging locations for emergency medical responders in a theme park to maximize the number of responses that can occur within a pre-specified time. The staging areas are selected from a candidate set of restricted access locations where the responders can store their equipment. Our solution approach considers all routes to access any park location, including areas that are unavailable to a regular guest. Theme parks are a highly dynamic environment. Because special events occurring in the park at certain hours (e.g., parades) might impact the responders' travel times, our model's decisions also include the time dimension in the location and re-location of the responders. Our solution provides the optimal location of the responders for each time partition, including backup responders. When an optimal solution is found, the model is also designed to consider alternate optimal solutions that provide a more balanced workload for the crews.
ContributorsLivingston, Noah Russell (Author) / Sefair, Jorge (Thesis director) / Askin, Ronald (Committee member) / Industrial, Systems and Operations Engineering Program (Contributor) / Barrett, The Honors College (Contributor)
Created2017-12