Artificial Intelligence and Robotics Commons^™

Toward A Theory-Based Natural Language Capability In Robots And Other Embodied Agents : Evaluating Hausser's Slim Theory And Database Semantics, Robin Kowalchuk Burk

Legacy Theses & Dissertations (2009 - 2024)

Computational natural language understanding and generation have been a goal of artificial intelligence since McCarthy, Minsky, Rochester and Shannon first proposed to spend the summer of 1956 studying this and related problems. Although statistical approaches dominate current natural language applications, two current research trends bring renewed focus on this goal. The nascent field of artificial general intelligence (AGI) seeks to evolve intelligent agents whose multi-subagent architectures are motivated by neuroscience insights into the modular functional structure of the brain and by cognitive science insights into human learning processes. Rapid advances in cognitive robotics also entail multi-agent software architectures that attempt …

Motivated Learning As An Extension Of Reinforcement Learning, Janusz Starzyk, Pawel Raif, Ah-Hwee Tan

Research Collection School Of Computing and Information Systems

We have developed a unified framework to conduct computational experiments with both learning systems: Motivated learning based on Goal Creation System, and reinforcedment learning using RL Q-Learning Algorithm. Future work includes combining motivated learning to set abstract motivations and manage goals with reinforcement learning to learn proper actions. This will allow testing of motivated learning on typical reinforcement learning benchmarks with large dimensionality of the state/action spaces.

Automated Interpretation Of The Tandem Mass Spectra Of Peptides Using Artificial Neural Networks, Timothy Patrick Manning

Theses

The manual interpretation of mass spectra is a complex and time consuming task. The problem of manually interpreting this data is further exacerbated by the large numbers of mass spectra which can potentially be produced in a single proteoniics experiment. This shows the need for high throughput approaches to the interpretation of mass spectra. Existing automated approaches are however error prone due to the complexity of the task. Accordingly, this thesis discusses and evaluates the application of neural networks to improving the sensitivity, specificity and robustness of current approaches to the automated interpretation of such mass spectral data.

Several neural …

A Boosting Framework For Visuality-Preserving Distance Metric Learning And Its Application To Medical Image Retrieval, Yang Liu, Rong Jin, Lily Mummert, Rahul Sukthankar, Adam Goode, Bin Zheng, Steven C. H. Hoi, Mahadev Satyanarayanan

Research Collection School Of Computing and Information Systems

Similarity measurement is a critical component in content-based image retrieval systems, and learning a good distance metric can significantly improve retrieval performance. However, despite extensive study, there are several major shortcomings with the existing approaches for distance metric learning that can significantly affect their application to medical image retrieval. In particular, "similarity" can mean very different things in image retrieval: resemblance in visual appearance (e.g., two images that look like one another) or similarity in semantic annotation (e.g., two images of tumors that look quite different yet are both malignant). Current approaches for distance metric learning typically address only one …

Periodic Resource Reallocation In Two-Echelon Repairable Item Inventory Systems, Hoong Chuin Lau, Jie Pan, Huawei Song

Research Collection School Of Computing and Information Systems

Given an existing stock allocation in an inventory system, it is often necessary to perform reallocation over multiple time points to address inventory imbalance and maximize availability. In this paper, we focus on the situation where there are two opportunities to perform reallocation within a replenishment cycle. We derive a mathematical model to determine when and how to perform reallocation. Furthermore, we consider the extension of this model to the situation allowing an arbitrary number of reallocations. Experimental results show that the two-reallocation approach achieves better performance compared with the single-reallocation approach found in the literature. We also illustrate how …

Inside The Selection Box: Visualising Active Learning Selection Strategies, Brian Mac Namee, Rong Hu, Sarah Jane Delany

Conference papers

Visualisations can be used to provide developers with insights into the inner workings of interactive machine learning techniques. In active learning, an inherently interactive machine learning technique, the design of selection strategies is the key research question and this paper demonstrates how spring model based visualisations can be used to provide insight into the precise operation of various selection strategies. Using sample datasets, this paper provides detailed examples of the differences between a range of selection strategies.

Svm Based Active Learning With Exploration, Patrick Lindstrom, Rong Hu, Sarah Jane Delany, Brian Mac Namee

Conference papers

No abstract provided.

Exploring The Frontier Of Uncertainty Space, Rong Hu, Patrick Lindstrom, Sarah Jane Delany, Brian Mac Namee

Conference papers

We aim to investigate methods balancing exploitation with exploration in active learning to improve the performance of uncertainty sampling. Two exploration guided sampling methods are compared to uncertainty sampling on various real-life datasets from the 2010 Active Learning Challenge. Our initial experiments seems to indicate that combining exploration with uncertainty sampling improves performance on certain datasets but not all.

Egal: Exploration Guided Active Learning For Tcbr, Rong Hu, Sarah Jane Delany, Brian Mac Namee

Conference papers

The task of building labelled case bases can be approached using active learning (AL), a process which facilitates the labelling of large collections of examples with minimal manual labelling effort. The main challenge in designing AL systems is the development of a selection strategy to choose the most informative examples to manually label. Typical selection strategies use exploitation techniques which attempt to refine uncertain areas of the decision space based on the output of a classifier. Other approaches tend to balance exploitation with exploration, selecting examples from dense and interesting regions of the domain space. In this paper we present …

Cognitive Effort For Multi Agent Systems, Luca Longo

Conference papers

Cognitive Effort is a multi-faceted phenomenon that has suffered from an imperfect understanding, an informal use in everyday life and numerous definitions. This paper attempts to clarify the concept, along with some of the main influencing factors, by presenting a possible heuristic formalism intended to be implemented as a computational concept, and therefore be embedded in an artificial agent capable of cognitive effort-based decision support. Its applicability in the domain of Artificial Intelligence and Multi-Agent Systems is discussed. The technical challenge of this contribution is to start an active discussion towards the formalisation of Cognitive Effort and its application in …

Prediction Of Brain Tumor Progression Using A Machine Learning Technique, Yuzhong Shen, Debrup Banerjee, Jiang Li, Adam Chandler, Yufei Shen, Frederic D. Mckenzie, Jihong Wang, Nico Karssemeijer (Ed.), Ronald M. Summers (Ed.)

Electrical & Computer Engineering Faculty Publications

A machine learning technique is presented for assessing brain tumor progression by exploring six patients' complete MRI records scanned during their visits in the past two years. There are ten MRI series, including diffusion tensor image (DTI), for each visit. After registering all series to the corresponding DTI scan at the first visit, annotated normal and tumor regions were overlaid. Intensity value of each pixel inside the annotated regions were then extracted across all of the ten MRI series to compose a 10 dimensional vector. Each feature vector falls into one of three categories:normal, tumor, and normal but progressed to …

Inversion Of 2d And 3d Dc Resistivity Imaging Data Forhigh Contrast Geophysical Regions Using Artificial Neuralnetworks., Ahmad Neyamadpour

Student Works (2010-2019)

In electrical resistivity imaging surveys, the field data along a profile are normally acquired as a subsurface distribution of apparent resistivity. One common method to obtain the true resistivity distribution is by inverting the apparent resistivity values. However, the inversion of DC resistivity imaging data is complex due to its non-linearity. This is especially true for regions with high resistivity contrast. For the complicated subsurface structure, especially when regions of high resistivity contrast exist, a conventional inversion technique based on least squares methods may not be able to invert the DC resistivity data with adequate accuracy. Therefore, in this study, …

Introducing Communication In Dis-Pomdps With Locality Of Interaction, Makoto Tasaki, Yuichi Yabu, Yuki Iwanari, Makoto Yokoo, Janusz Marecki, Pradeep Reddy Varakantham, Milind Tambe

Research Collection School Of Computing and Information Systems

The Networked Distributed POMDPs (ND-POMDPs) can model multiagent systems in uncertain domains and has begun to scale-up the number of agents. However, prior work in ND-POMDPs has failed to address communication. Without communication, the size of a local policy at each agent within the ND-POMDPs grows exponentially in the time horizon. To overcome this problem, we extend existing algorithms so that agents periodically communicate their observation and action histories with each other. After communication, agents can start from new synchronized belief state. Thus, we can avoid the exponential growth in the size of local policies at agents. Furthermore, we introduce …