Program
At a glance
All events take place at the 
Lister Center, unless noted otherwise.
|
Tuesday
May 16, 2017 8:30—10:00 Graduate Student Symposium: Session 1
10:00—10:30 Coffee break
10:30—12:00 Graduate Student Symposium: Session 2
12:00—12:30 M.Sc. Award Winner Talk
12:30—14:00 Lunch break
14:00—17:30 Tutorials (incl. 30min break)
17:30—18:30 NSERC Presentation
|
Wednesday
May 17, 2017 08:00—8:30 Breakfast
08:30—9:00 Joint Welcoming Session
9:00—10:30 Session 1: Data Mining and Machine Learning
10:30—11:00 Coffee break
11:00—12:30 Session 2: Planning and Combinatorial Optimization
12:30—14:00 Lunch break
14:00—15:00 Session 3 - Keynote Speaker: Hugo Larochelle
15:00—15:30 Coffee break
15:30—17:00 Session 4: AI Applications
17:00—18:00 Poster Session for Short Papers
18:00—20:00 Welcome Reception at Congress Centre
|
Thursday
May 18, 2017 8:00—9:00 Breakfast
9:00—10:30 Session 5: Natural Language Processing
10:30—11:00 Coffee break
11:00—12:30 Session 6: Uncertainty and Preference Reasoning
12:30—14:00 Lunch break
14:00—15:00 Session 7 - Keynote speaker: Adnan Darwiche
15:00—15:30 Coffee break
15:30—17:00 Session 8: Data Mining and Machine Learning
17:00—18:00 Poster Session for Short Papers
18:00—19:00 CAIAC AGM
20:00—23:00 Awards Banquet, at Fairmont MacDonald
|
Friday
May 19, 2017 8:00—8:40 Breakfast
8:40—10:00 Session 9: Agent Systems
10:00—10:30 Coffee break
10:30—11:30 Session 10 - Keynote Speaker: Robert Holte
11:30—12:30 Session 11 - PhD Award Winner
12:30—14:00 Lunch - AI/GI/CRV steering committee meeting
14:00—15:30 Session 12 - Industry Track Presentations
15:30—16:00 Coffee break
16:00—16:15 Closing Remarks
|
Detailed Program
Long paper presentation (AI-LP#): 20min (incl. 5 min for Q&A)
Short paper presentation (AI-SP#): 10min (questions during the poster session)
Invited talk: 1 hour (incl. 15min for Q&A)
GSS paper presentation (GSS-P#): 30min (incl. 10 min for Q&A)
Tuesday, May 16, 2017
8:30-10:00 Graduate Student Symposium: Session 1
GSS-P6Diverse Action Costs in Heuristic Search and Planning
GSS-P3Collaborative Filtering with Users' Qualitative and Conditional Preferences
GSS-P7Deep Multi-Cultural Graph Representation Learning
10:00-10:30 Coffee break
10:30-12:00 Graduate Student Symposium: Session 2
GSS-P9Who Is the Artificial Author?
GSS-P5Machine Learning Techniques to Unveil and Understand Pseudomonas aeruginosa Survival Mechanism in Nutrient Depleted Water
GSS-P4Identification and Classification of Alcohol-Related Violence in Nova Scotia Using Machine Learning Paradigms
12:00-12:30 CAIAC 2017 Masters Thesis Award Winner Talk
12:30-14:00 Lunch break
14:00-17:30 Tutorials (including a 30 minute break)
T1: The Evolution of Darwinian Networks
Aurora Room
Presenters: Cory J. Butz, Jhonatan S. Oliveira and André E. dos Santos, University of Regina
This tutorial provides an overview of the development of Darwinian networks (DNs). DNs simplify working with Bayesian networks (BNs). DNs adapt a handful of well-known concepts in biology into a single framework that is surprisingly simple, yet remarkably robust. With respect to modeling, on one hand, DNs not only represent BNs, but also faithfully represent the testing of independencies in a more straightforward fashion. On the other hand, with respect to three exact inference algorithms in BNs, DNs simplify each of them, while unifying all of them. DNs can determine good elimination orderings using the same terminology as used for modeling and inference. Practical benefits of DNs include faster algorithms for inference and modeling.
T2: Tools for Natural Language Processing
Maple Leaf Room
Presenters: Greg Kondrak, Garrett Nicolai, and Bradley Hauer, University of Alberta
The tutorial is aimed at graduate students and other AI 2017 participants that are interested in incorporating Natural Language Processing (NLP) techniques into their research. After an introduction to the foundations of NLP, the tutorial will focus on open-source toolkits, such as NLTK, Stanford CoreNLP, DirecTL+, and Word2Vec. Installation advice, usage guidelines, and illustrative examples will be provided. We will discuss how to apply those tools to various NLP tasks, such as part-of-speech tagging, named-entity recognition, parsing, lemmatizing, and stemming. The approach will be “hands-on”, enabling the participants to put the newly acquired skills into immediate practice. The tutorial will conclude by demonstrating several NLP applications and the underlying data sets. No previous NLP background is required.
T3: Accelerating Data Analytics: Keeping Pace with Business Demands
Room: ATH 332
Presenters: Behrouz Far and Emad Mohamed, University of Calgary
The volume of data an organization has is less important than how the datasets are being used to generate some applicable ideas that can improve business. Substantial data growth, different data warehouses, and different data format are hindering organizations to provide a better understanding of their business domain due to the amount of efforts consumed in adopting the data into a common usable data model for processing. Data warehouse, data science, and big data analytics tools are substantially used in operation research, recommendation systems, healthcare systems and personalized health outcome improvement, etc., and become increasingly operational in nature. However, this means more data marts, more preprocessing steps, and a wider reach throughout organizations. Keeping pace with this evolution requires designing of predictive analytics models that provide quantifiable and actionable insights to improve a specific business domain. To this end, organizations start by targeting an innovative answer to a business. To find this answer in a timely manner, organizations require to get their first data mining project off the ground and prove the value of data mining to their organization “Proof-of-Concept”.
The data science initiative at the University of Calgary is one of the research priority themes, and we lead this initiative at the Schulich School of Engineering, Electrical and Computer Engineering Department. At Winter-2015 we had the opportunity to design and deliver the first undergrad course at the University of Calgary “Engineering Large-Scale Analytics Systems” (SENG-501.5), where the concepts of big data analytics were discussed on the theoretical and practical aspects. Recently, we had the opportunity to build the Multi-Modal Data Fusion (MMDF) lab at the department. The lab is equipped by a Hadoop and Spark cluster on a top of a commodity hardware. The main purpose of the lab is to make it available for various types of data science projects that utilize big volume of data for designing proof of concept prototypes in different business domains, e.g., autonomous vehicle, healthcare analytics, etc.
In this tutorial, we will share with the audience our experience in the following:
- A step-by-step guide for building and installing an Apache Hadoop and Spark cluster of a commodity hardware using the IBM Open Data Platform (IODP).
- Installing R and H2O cloud on top of the cluster for machine learning and deep learning tasks
- Explain the MapRedcue and Spark frameworks for batch and in-memory analytics.
- Illustrate how to build a data science project for non-technical “programmer” audience using Data Science Studio “Community Edition” Software.
- Explain the limitation of the MapRedcue and Spark frameworks and the suitable types of problems.
- Discuss and explain our “Success Stories” in building data science projects and our partners “Alberta Health Services” and Shell Canada.
- Demo: if the bandwidth allows, we will remotely access the UofC MMDF cluster to present many practical use-cases.
17:30-18:30 NSERC Presentation
Wednesday, May 17, 2017
08:00-8:30 Breakfast
08:30-9:00 Joint Welcoming Session - Conference Opening
9:00-10:30 Session 1: Data Mining and Machine Learning
Chair: Nathalie Japkowicz, American University
AI-LP10On Generalized Bellman Equations and Temporal-Difference Learning
AI-LP45Person Identification Using Discriminative Visual Aesthetic
AI-LP57A global search approach for inducing Oblique Decision Trees using Differential Evolution
AI-SP15Improving Active Learning for One-Class Classification using Dimensionality Reduction
AI-SP40Local and Global Influence on Twitter
AI-SP46The Impact of Toxic Language on the Health of Reddit Communities
10:30-11:00 Coffee break
11:00-12:30 Session 2: Planning and Combinatorial Optimization
Chair: Howard Hamilton, University of Regina
AI-LP3A Worst-Case Analysis of Constraint-Based Algorithms for Exact Multi-Objective Combinatorial Optimization
AI-LP27Metaheuristics for Score-and-Search Bayesian Network Structure Learning
AI-SP18SmartHome Energy Saving Using a Multi Objective Approach Based on Appliances Usage Profiles
AI-SP59Sequence-based Bidirectional Merge Map-Matching Algorithm for Simplified Road Network
AI-SP43On the Role of Possibility in Action Execution and Knowledge in the Situation Calculus
AI-SP28Combinatorial Reverse Electricity Auctions
AI-SP61Policy Conflict Resolution in IoT via Planning
12:30-14:00 Lunch break
14:00-15:00 Session 3 - Keynote Speaker: Hugo Larochelle
Chair: Philippe Langlais, Université de Montréal
Autoregressive Generative Models with Deep Learning
15:00-15:30 Coffee break
15:30-17:00 Session 4: AI Applications
Chair: Marina Sokolova, University of Ottawa
AI-LP5Learning Physical Properties of Objects Using Gaussian Mixture Models
AI-LP33Design and Implementation of a Smart Quotation System
AI-LP37Knowledge Discovery in Graphs through Vertex Separation
AI-LP60A Knowledge Acquisition System for Price Change Rules
AI-SP50Automatic Household Identification for Historical Census Data
17:00 - 18:00 Poster Session for Short Papers
18:00-21:00 Reception
Thursday, May 18, 2017
8:00-9:00 Breakfast
9:00-10:30 Session 5: Natural Language Processing
Chair:Diana Inkpen, University of Ottawa
AI-LP12Matrix Models with Feature Enrichment for Relation Extraction
AI-LP52Domain Adaptation for Detecting Mild Cognitive Impairment
AI-LP44Speech Intention Classification with Multimodal Deep Learning
AI-SP2Layerwise Interweaving Convolutional LSTM
AI-SP23Using Cognitive Computing to Get Insights on Personality Traits from Twitter Messages
AI-SP38Confused and Thankful: Multi-Label Sentiment Classification of Health Forums
10:30-11:00 Coffee break
11:00-12:30 Session 6: Uncertainty and Preference Reasoning
Chair: Xin Wang, University of Calgary
AI-LP1Probabilistic TCP-Net
AI-LP30Resolving Inconsistencies of Scope Interpretations in Sum-Product Networks
AI-LP34A Sparse Probabilistic Model of User Preference Data
AI-SP31On Converting Sum-Product Networks into Bayesian Networks
AI-SP11Fuzzy Rough Set-based Unstructured Text Categorization
AI-SP56Bayesian Networks to Model Pseudomonas aeruginosa Survival Mechanism and Identify Low Nutrient Response Genes in Water
12:30-14:00 Lunch break
14:00-15:00 Session 7 - Keynote speaker: Adnan Darwiche
Chair: Cory Butz, University of Regina
Tractable Learning in Structured Probability Spaces
15:00-15:30 Coffee break
15:30-17:00 Session 8: Data Mining and Machine Learning
Chair: Samira Sadaoui, University of Regina
AI-SP16Somatic Copy Number Alteration-Based Prediction of Molecular Subtypes of Breast Cancer Using Deep Learning Model
AI-SP53An Improved Data Sanitization Algorithm for Privacy Preserving Medical Data Publishing
AI-SP24Reflexive Regular Equivalence for Bipartite Data
AI-SP48Investigating Citation Linkage with Machine Learning
AI-SP14Classification of Imbalanced Auction Fraud Data
AI-SP41Multi-label Learning Through Minimum Spanning Tree-based Subset Selection and Feature Extraction
AI-SP4Comparative Study of Dimensionality Reduction Methods Using Reliable Features for Multiple Datasets Obtained by rs-fMRI in ADHD Prediction
AI-SP20Time-Dependent Smart Data Pricing Based on Machine Learning
AI-SP55Time Prediction of the Next Refueling Event: A Case Study
17:00 - 18:00 Poster Session for Short Papers
18:00-19:00 CAIAC AGM
20:00-23:00 Awards Banquet
Friday, May 19, 2017
8:00-8:40 Breakfast
8:40-10:00 Session 9: Agent Systems
Chair: Gabriel Murray, University of the Fraser Valley
AI-LP6Quantified Coalition Logic of Knowledge, Belief and Certainty
AI-LP9Modelling Personality-based Individual Differences in the Use of Emotion Regulation Strategies
AI-LP17Stoic Ethics for Artificial Agents
AI-LP42Active Team Management Strategies for Multi-Robot Teams in Dangerous Environments
10:00-10:30 Coffee break
10:30-11:30 Session 10 - Keynote Speaker: Robert Holte
Chair: Peter van Beek, University of Waterloo
Heuristic Search: Something Old and Something New
11:30-12:30 Session 11 - PhD Award Winner
12:30-14:00 Lunch - AI/GI/CRV steering committee meeting
14:00-15:30 Session 12 - AI in Industry, Challenges in Real Life Products
Chair: Éric Charton, Montréal Yellow Pages
14:05-14:25: Identifying news from social media and assessing its credibility
Presenter : Sameena Shah, director research, Thomson Reuters
News moves markets. Increasingly we find news breaking first on Twitter. However, discovering news events from amongst social media's mundane conversations is a big challenge. This gets further complicated by the presence of questionable and fake news. Sameena and her team created Reuters Tracer, a machine able to detect and verify news events at scale and speed. Sameena will speak about some of the machine learning that went into Reuters Tracer and how it being used in the newsroom.
14:25-14:45: Agricultural Decision Support using HYDRA: a semantic data federation engine
Presenter : Christopher J. O. Baker, IPSNP Computing Inc.
HYDRA is a semantic data federation designed for subject matter experts with minimal technical expertise. Using a graphical query canvas consultants and business analysts can construct complex queries using a series of simple keywords or key-phrases. Completed graphical queries are translated to SPARQL and issued to HYDRA. Employing a registry of SADI semantic web services the HYDRA is able to automatically discover services, plan and execute complex workflows according to an analyst’s ad-hoc query needs. In this demonstration consultants use HYDRA to select crop varieties, dynamically compute profit margins, and determine appropriate pesticides for each crop variety.
14:45-15:05: Choosing the AI technology to fit the product need not the ego
Presenter : Eric Charton, Montreal Yellow Pages
Yellow Pages digital properties are serving millions of local search queries every day. At various stage of the product development, information retrieval related algorithms, relevance feedback loop, and metrics production, AI and machine learning technologies can be involved. But in real life products, there is always a need to arbitrate between scalability, usefulness and return on investment. In this presentation, we will describe three applications of AI based on neural networks, Markov Logic Network but also on more simple and flexible rule based systems. We will explain why sometimes pragmatic choices can outperform sophisticated technologies.
15:05-15:30: Round table: AI in industry, challenges in real life products
A lively discussion and exchange between speakers and the public about AI challenges in industry.
