Physical Sciences and Mathematics Commons^™

Using Machine Learning To Predict Chemotherapy Response In Cell Lines And Patients Based On Genetic Expression, Dimo Angelov

Electronic Thesis and Dissertation Repository

The goal of this thesis was to examine different machine learning techniques for predicting chemotherapy response in cell lines and patients based on genetic expression. After trying regression, multi-class classification techniques and binary classification it was concluded that binary classification was the best method for training models due to the limited size of available cell line data. We found support vector machine classifiers trained on cell line data were easier to use and produced better results compared to neural networks. Sequential backward feature selection was able to select genes for the models that produced good results, however the greedy algorithm …

The Wonders Of The Spreadsheet Tool For Data Management And Insights, Michelle L. F. Cheong

Research Collection School Of Computing and Information Systems

Ask any student at the Singapore Management University (SMU) toname one of the most practical and useful courses offered by theuniversity. The answer would inevitably include CAT. CAT stands forthe "Computer as an Analysis Tool" course. Originally based on acourse of the same title offered by the Wharton Business School, thefocus of CAT was shifted to provide business students the essentialpractical skills and necessary “real-world” exposure to better usepersonal computers for resolving business problems. The course isbasically centred on using the Excel spreadsheet to work onambiguous ill-defined problems (Leong & Cheong, 2009). Over theyears, three editions of a textbook have …

Privacy In Context-Aware Mobile Crowdsourcing Systems, Thivya Kandappu, Archan Misra, Shih-Fen Cheng, Hoong Chuin Lau

Research Collection School Of Computing and Information Systems

Mobile crowd-sourcing can become as a strategy to perform time-sensitive urban tasks (such as municipal monitoring and last mile logistics) by effectively coordinating smartphone users. The success of the mobile crowd-sourcing platform depends mainly on its effectiveness in engaging crowd-workers, and recent studies have shown that compared to the pull-based approach, which relies on crowd-workers to browse and commit to tasks they would want to perform, the push-based approach can take into consideration of worker’s daily routine, and generate highly effective recommendations. As a result, workers waste less time on detours, plan more in advance, and require much less planning …

Feature Learning Via Partial Differential Equation With Applications To Face Recognition, Cong Fang, Zhenyu Zhao, Pan Zhou, Zhouchen Lin

Research Collection School Of Computing and Information Systems

Feature learning is a critical step in pattern recognition, such as image classification. However, most of the existing methods cannot extract features that are discriminative and at the same time invariant under some transforms. This limits the classification performance, especially in the case of small training sets. To address this issue, in this paper we propose a novel Partial Differential Equation (PDE) based method for feature learning. The feature learned by our PDE is discriminative, also translationally and rotationally invariant, and robust to illumination variation. To our best knowledge, this is the first work that applies PDE to feature learning …

Directed Acyclic Graph Continuous Max-Flow Image Segmentation For Unconstrained Label Orderings, John Sh Baxter, Martin Rajchl, A. Jonathan Mcleod, Jing Yuan, Terry M. Peters

Robarts Imaging Publications

Label ordering, the specification of subset–superset relationships for segmentation labels, has been of increasing interest in image segmentation as they allow for complex regions to be represented as a collection of simple parts. Recent advances in continuous max-flow segmentation have widely expanded the possible label orderings from binary background/foreground problems to extendable frameworks in which the label ordering can be specified. This article presents Directed Acyclic Graph Max-Flow image segmentation which is flexible enough to incorporate any label ordering without constraints. This framework uses augmented Lagrangian multipliers and primal–dual optimization to develop a highly parallelized solver implemented using GPGPU. This …

Exploring Algorithms To Recognize Similar Board States In Arimaa, Malik Khaleeque Ahmed

Theses and Dissertations

The game of Arimaa was invented as a challenge to the field of game-playing artificial intelligence, which had grown somewhat haughty after IBM's supercomputer Deep Blue trounced world champion Kasparov at chess. Although Arimaa is simple enough for a child to learn and can be played with an ordinary chess set, existing game-playing algorithms and techniques have had a difficult time rising up to the challenge of defeating the world's best human Arimaa players, mainly due to the game's impressive branching factor. This thesis introduces and analyzes new algorithms and techniques that attempt to recognize similar board states based on …

Decentralized Planning In Stochastic Environments With Submodular Rewards, Rajiv Ranjan Kumar, Pradeep Varakantham, Akshat Kumar

Research Collection School Of Computing and Information Systems

Decentralized Markov Decision Process (Dec-MDP) providesa rich framework to represent cooperative decentralizedand stochastic planning problems under transition uncertainty.However, solving a Dec-MDP to generate coordinatedyet decentralized policies is NEXP-Hard. Researchershave made significant progress in providing approximate approachesto improve scalability with respect to number ofagents. However, there has been little or no research devotedto finding guarantees on solution quality for approximateapproaches considering multiple (more than 2 agents)agents. We have a similar situation with respect to the competitivedecentralized planning problem and the StochasticGame (SG) model. To address this, we identify models in thecooperative and competitive case that rely on submodular rewards,where we show …

Dynamic Repositioning To Reduce Lost Demand In Bike Sharing Systems, Supriyo Ghosh, Pradeep Varakantham, Yossiri Adulyasak, Patrick Jaillet

Research Collection School Of Computing and Information Systems

Bike Sharing Systems (BSSs) are widely adopted in major cities of the world due to concerns associated with extensive private vehicle usage, namely, increased carbon emissions, traffic congestion and usage of nonrenewable resources. In a BSS, base stations are strategically placed throughout a city and each station is stocked with a pre-determined number of bikes at the beginning of the day. Customers hire the bikes from one station and return them at another station. Due to unpredictable movements of customers hiring bikes, there is either congestion (more than required) or starvation (fewer than required) of bikes at base stations. Existing …

Toward A Collaborative Ai Framework For Assistive Dementia Care, Tze-Yun Leong

Research Collection School Of Computing and Information Systems

We envision an integrated framework for supporting the development and deployment of human-aware, general artificial intelligence (AI) that needs to collaborate in uncertain, changing environments. We examine the technology and system requirements of building assistive care agents for dementia or cognitive impaired patients through the continuum of care. We summarize the new AI capabilities and show examples of how an evolving, adaptive development approach would be able to support the basic functionalities and applications in a sound, practical, and scalable manner. We highlight the challenges and the opportunities involved in realizing the proposed framework, and call for future research and …

Optimizing Expectation With Guarantees In Pomdps, Krishnendu Chatterjee, Guillermo A. Perez, Jean-François Raskin, Dorde Zikelic

Research Collection School Of Computing and Information Systems

A standard objective in partially-observable Markov decision processes (POMDPs) is to find a policy that maximizes the expected discounted-sum payoff. However, such policies may still permit unlikely but highly undesirable outcomes, which is problematic especially in safety-critical applications. Recently, there has been a surge of interest in POMDPs where the goal is to maximize the probability to ensure that the payoff is at least a given threshold, but these approaches do not consider any optimization beyond satisfying this threshold constraint. In this work we go beyond both the "expectation" and "threshold" approaches and consider a "guaranteed payoff optimization (GPO)" problem …

An Efficient Approach To Model-Based Hierarchical Reinforcement Learning, Zhuoru Li, Akshay Narayan, Tze-Yun Leong

Research Collection School Of Computing and Information Systems

We propose a model-based approach to hierarchical reinforcement learning that exploits shared knowledge and selective execution at different levels of abstraction, to efficiently solve large, complex problems. Our framework adopts a new transition dynamics learning algorithm that identifies the common action-feature combinations of the subtasks, and evaluates the subtask execution choices through simulation. The framework is sample efficient, and tolerates uncertain and incomplete problem characterization of the subtasks. We test the framework on common benchmark problems and complex simulated robotic environments. It compares favorably against the stateof-the-art algorithms, and scales well in very large problems.

Collective Multiagent Sequential Decision Making Under Uncertainty, Duc Thien Nguyen, Akshat Kumar, Hoong Chuin Lau

Research Collection School Of Computing and Information Systems

Multiagent sequential decision making has seen rapid progress with formal models such as decentralized MDPs and POMDPs. However, scalability to large multiagent systems and applicability to real world problems remain limited. To address these challenges, we study multiagent planning problems where the collective behavior of a population of agents affects the joint-reward and environment dynamics. Our work exploits recent advances in graphical models for modeling and inference with a population of individuals such as collective graphical models and the notion of finite partial exchangeability in lifted inference. We develop a collective decentralized MDP model where policies can be computed based …

Recurrent Neural Networks With Auxiliary Labels For Cross-Domain Opinion Target Extraction, Ying Ding, Jianfei Yu, Jing Jiang

Research Collection School Of Computing and Information Systems

Opinion target extraction is a fundamental task in opinion mining. In recent years, neural network based supervised learning methods have achieved competitive performance on this task. However, as with any supervised learning method, neural network based methods for this task cannot work well when the training data comes from a different domain than the test data. On the other hand, some rule-based unsupervised methods have shown to be robust when applied to different domains. In this work, we use rule-based unsupervised methods to create auxiliary labels and use neural network models to learn a hidden representation that works well for …

A Riemannian Network For Spd Matrix Learning, Zhiwu Huang, Gool L. Van

Research Collection School Of Computing and Information Systems

Symmetric Positive Definite (SPD) matrix learning methods have become popular in many image and video processing tasks, thanks to their ability to learn appropriate statistical representations while respecting Riemannian geometry of underlying SPD manifolds. In this paper we build a Riemannian network architecture to open up a new direction of SPD matrix non-linear learning in a deep model. In particular, we devise bilinear mapping layers to transform input SPD matrices to more desirable SPD matrices, exploit eigenvalue rectification layers to apply a non-linear activation function to the new SPD matrices, and design an eigenvalue logarithm layer to perform Riemannian computing …

Malware Detection Using The Index Of Coincidence, Bhavna Gurnani

Master's Projects

In this research, we apply the Index of Coincidence (IC) to problems in malware analysis. The IC, which is often used in cryptanalysis of classic ciphers, is a technique for measuring the repeat rate in a string of symbols. A score based on the IC is applied to a variety of challenging malware families. We nd that this relatively simple IC score performs surprisingly well, with superior results in comparison to various machine learning based scores, at least in some cases.

Deep Learning Methods For Protein Torsion Angle Prediction, Haiou Li, Jie Hou, Badri Adhikari, Qiang Lyu, Jianlin Cheng

Badri Adhikari

No abstract provided.

Imitating The Brain: Autonomous Robots Harnessing The Power Of Artificial Neural Networks, Mohammad Khan

Computer Science Honors Papers

Artificial Neural Networks (ANNs) imitate biological neural networks, which can have billions of neurons with trillions of interconnections. The first half of this paper focuses on fully-connected ANNs and hardware neural networks. The latter half of this paper focuses on Deep Learning, a strategy in Artificial Intelligence based on massive ANN architectures. We focus on Deep Convolutional Neural Networks, some of which are capable of differentiating between thousands of objects by self-learning from millions of images. We complete research in two areas of focus within the field of ANNs, and we provide ongoing work for and recommend two more areas …

Ai Education: Open-Access Educational Resources On Ai, Todd W. Neller

Computer Science Faculty Publications

Open-access AI educational resources are vital to the quality of the AI education we offer. Avoiding the reinvention of wheels is especially important to us because of the special challenges of AI Education. AI could be said to be “the really interesting miscellaneous pile of Computer Science”. While “artificial” is well-understood to encompass engineered artifacts, “intelligence” could be said to encompass any sufficiently difficult problem as would require an intelligent approach and yet does not fall neatly into established Computer Science subdisciplines. Thus AI consists of so many diverse topics that we would be hard-pressed to individually create quality learning …

Ai Education: Deep Neural Network Learning Resources, Todd W. Neller

Computer Science Faculty Publications

In this column, we focus on resources for learning and teaching deep neural network learning. Many exciting advances have been made in this area of late, and so many resources have become available online that the flood of relevant concepts and techniques can be overwhelming. Here, we hope to provide a sampling of high-quality resources to guide the newcomer into this booming field. [excerpt]

Ai Education: Machine Learning Resources, Todd W. Neller

Computer Science Faculty Publications

In this column, we focus on resources for learning and teaching three broad categories of machine learning (ML): supervised, unsupervised, and reinforcement learning. In ournext column, we will focus specifically on deep neural network learning resources, so if you have any resource recommendations, please email them to the address above. [excerpt]

Performance Verification For Robot Missions In Uncertain Environments, Damian Lyons, Ron Arkin, Shu Jiang, Matt O'Brien, Feng Tang, Peng Tang

Faculty Publications

Abstract—Certain robot missions need to perform predictably in a physical environment that may have significant uncertainty. One approach is to leverage automatic software verification techniques to establish a performance guarantee. The addition of an environment model and uncertainty in both program and environment, however, means the state-space of a model-checking solution to the problem can be prohibitively large. An approach based on behavior-based controllers in a process-algebra framework that avoids state-space combinatorics is presented here. In this approach, verification of the robot program in the uncertain environment is reduced to a filtering problem for a Bayesian Network. Validation results …

Petrochemical Plant Console Operator Workload:The Issues, David A. Strobhar

H-Workload 2017: Models and Applications (Works in Progress)

The console operators of certain petrochemical processes must maintain high levels of performance during process upsets or endanger personnel safety and the environment. Mismanagement of an upset can result in explosions, fires, and the release of hazardous chemicals to the environment. The change in workload from steady state to upset operation is significant, with alarms and control changes that are of an order of magnitude. This paper describes the state of console activity in process plants, particularly the increase with key upsets. Quantitative data on the nature of the console operator’s position, its workload during normal operation, and the requirements …

A Workload-Centered Perspective On Reduced Crew Operations In Commercial Aviation, Daniela Schmid

H-Workload 2017: Models and Applications (Works in Progress)

Mental workload of a pilot, in short workload, depends on various characteristics of different accumulated tasks on the flight deck. Exogenous task demands and endogenous supply of attentional or information processing resources determine workload [1]. Performance is expect to drop if the demand exceeds the available resources of the pilot. Expertise and experience modulate the endogenous sup- ply of resources like perceiving, updating memory, planing, making a decision, and executing and processing a response. Subsequently, workload manifests in performance variables, subjective experience, and physiological parameters [2]. This is how we can summarize workload very brie y to introduce a model …

A System To Monitor Cognitive Workload In Naturalistic High-Motion Environments, Bethany K. Bracken, Seth Elkin-Frankston, Noa Palmon, Michael Farry, Blaise De B Frederick

H-Workload 2017: Models and Applications (Works in Progress)

Across many careers, individuals face alternating periods of high and low attention and cognitive workload can impair cognitive function and undermine job performance. We have designed and are developing an unobtrusive system to Monitor, Extract, and Decode Indicators of Cognitive Workload (MEDIC) in naturalistic, high-motion environments. MEDIC is designed to warn individuals, teammates, or supervisors when steps should be taken to augment cognitive readiness. We first designed and manufactured a forehead sensor device that includes a custom fNIRS sensor and a three-axis accelerometer designed to be mounted on the inside of a baseball cap or headband, or standard issue gear …

Smart Workload Balancing, Ferdinand Coster

H-Workload 2017: Models and Applications (Works in Progress)

The cognitive workload of operators working with automated systems should neither be too high nor too low. A static level of automation is unable to cope with systems that produce large fluctuations in cognitive workload, therefore a method for adaptive automation is proposed that could balance workload by intelligently choosing what to automate and when. To this end the concept of the Cognitive Workload Value factor is introduced, which takes into account both workload and situation awareness. This initial work introduces a possible framework for categorizing and using different workload and situation awareness measures.

Towards A Not Obtrusive Low Cost Biosystem To Assess Risk Perception In Workplace Through Stress Detection, Emanuele Bellini, Serena Benevenuti, Chiara Batistini

H-Workload 2017: Models and Applications (Works in Progress)

The main aim of the article is to build a method to assess risk perception in real time in order to early detect and prevent risk behaviors and possible human errors. To this end, the relation between mental workload and stress as critical factors affecting risk perception has been investigated. In particular the mental-physical activation generated by an increment of the workload has the effect of reducing the resources needed to perceive risk increasing the worker vulnerability. The complexity of the stress phenomenon suggested the adoption of an integrated view. The Functional Model has been adopted to for its holistic …

Reducing Peak Workload In The Cockpit: A Human In The Loop Simulation Evaluating New Runway Selection Tool, Tanja Bos, Rolf Zon, Wilfred Rouwhorst

H-Workload 2017: Models and Applications (Works in Progress)

In efforts to increase safety and reduce peak workload situations in the cockpit, a tool with a different interaction style was developed for use in case of a runway change instructed by Air Traffic Control during approach. In an experiment a workload comparison was made between the new tool and the conventional cockpit. Workload was measured by means of a self-rating after each experiment run, as well as eye blink frequency during each run. Results show that the self-rated workload decreases with the new tool for one of the two crew members and the blink frequency suggests a workload decrease …

Smart Homes Enhance Seniors’ Safety, Singapore Management University

Research@SMU: Connecting the Dots

Professor Tan Hwee Pink and researchers at iCity Lab are using sensors to increase the safety of seniors who live independently in their own homes.

See the papers:

• Online detection of behavioral change using unobtrusive eldercare monitoring system
• Improving the sensitivity of unobtrusive inactivity detection in sensor-enabled homes for the elderly
• SHINESeniors: Personalized services for active ageing-in-place

Complex Affect Recognition In The Wild, Behnaz Nojavanasghari

Electronic Theses and Dissertations

Artificial social intelligence is a step towards human-like human-computer interaction. One important milestone towards building socially intelligent systems is enabling computers with the ability to process and interpret the social signals of humans in the real world. Social signals include a wide range of emotional responses from a simple smile to expressions of complex affects. This dissertation revolves around computational models for social signal processing in the wild, using multimodal signals with an emphasis on the visual modality. We primarily focus on complex affect recognition with a strong interest in curiosity. In this dissertation,we ?rst present our collected dataset, EmoReact. …

Establishing A-Priori Performance Guarantees For Robot Missions That Include Localization Software, Damian Lyons, Ron Arkin, Shu Jiang, Matt O'Brien, Feng Tang, Peng Tang

Faculty Publications

One approach to determining whether an automated system is performing correctly is to monitor its performance, signaling when the performance is not acceptable; another approach is to automatically analyze the possible behaviors of the system a-priori and determine performance guarantees. Thea authors have applied this second approach to automatically derive performance guarantees for behaviorbased, multi-robot critical mission software using an innovative approach to formal verification for robotic software. Localization and mapping algorithms can allow a robot to navigate well in an unknown environment. However, whether such algorithms enhance any specific robot mission is currently a matter for empirical validation. Several …