ECE Machine Learning Lab

The University of Waterloo ECE Machine Learning Lab (UWECEML) carries out work on a variety of topics within Artificial Intelligence and Machine Learning with a focus on using real world problems to discover computationally hard problems of modelling uncertainty, learning predictive models and enabling decision making. Below are some of the topics, domains and concepts of ongoing and prior work in the lab.


AI/ML Topics

Deep Learning

Convolutional Neural Networks (CNNs) are a type of multi-stage, deep neural network which use alternating layers of convolutions, to correlate nearby points, and pooling, to reduce the data size to encourage abstraction. We are exploring the ability of CNNs to perform more powerful reasoning in spatiotemporal doamins like forest fire managemnet and brain imaging.

Parameter Tuning

The main Idea is to treat an entire computer program as a function which has some input parameters, settings, tuning variables, or thresholds. The machine learning problem is to find the best configuration parameters in such a way that the program maximizes some metric such as computational time or the accuracy or quality of the output. For some of our domains the complex program that requires tuning is a simulator of an ecosystem, disease spread or forest fire. (YAH) From real weather satellite forest data. There area natural Metrics for measuring the accuracy of Simulation run or the Speed of the complexity of it. We can Tuning parameters of the simulator in that way The challenge Here is the do it Based on real data and to create general algorithms which can tun general types of simulations To match that input data.

Anomaly Detection

Anomaly detection is the problem identifying within a given dataset a collect of events or datapoints that lie outside the “normal” behaviour for the system. This is similar to outlier detection in regression analysis and can be thought of as a two class, unsupervised clustering problem where one class contains normal datapoints and the other contains the anomalies. We may know something about the anomalies ahead of time or, more often, we will need to learn to identify them by their rarity or their dissimilarity with the vast majority of normal datapoints. In the driving domain the most important anomalies are unsafe driving situations. 

In computational sustainability problems in ecology, climate science, migration modelling and forest management an important class of anomalous situations are catastrophic changes to the system. These could be very large forest fires which reach a destructively high temperature; or they could be a widespread increase in mortality or fertility in plants or fish; or it could be an average yearly temperature beyond an historical threshold. In all these cases the catastrophic event could be recognized as an anomaly when looking back on the data, but looking forward we would need to predict the imminent change in class or state that results in the anomaly. This problem of predictive clustering and catastrophe detection/prediction is one of the longer terms goals of my research in several areas.


Optimization and Decision Making Under Uncertainty

This includes work on Reinforcement Learning using direct policy search as well as exact solution to Markov Decision Processes with confidence bounds particularly in complex domains especially when there are vary large state and action spaces. A wide range of optimization algorithms have been developed that are each appropriate in different types of problems. Equilibrium Policy Gradients utilize a parametrized policy defined as the equilibrium distribution of a Markov chain built from local causal policies. We have used this algorithm to deal with the complexity caused by spatial structure common in computational sustainability domains. There is also work on MDP planning algorithms with provable confidence intervals upon completion and other Reinforcement Learning approaches.

Graphical Models

Some of out work looks at modelling, inference and learning of interpretable probabilistic graphical models. Of particularl interest are cases where the underlying data is relational, such as social networks, or there is a need for both causal and correlational structure, cyclic structure in other words, within the same model. This arises naturally when trying to represent spatial policies using graphical models but cyclic structure can arise in many relational domains and existing fully directed or fully undirected models have trouble dealing with it.


Data/Application Domains

  • Medical Imaging - learning predictive classification models for diffusion MRI data to provide decision support for degenerative brain diseases

  • Forest Magagement - optimization of sustainablable harvest policies under spatial constraints, modelling of decision making about treatment against forest wildfires using Markov decision processes, learning dynamics of forest fire spread from simulations and satellite data, dynamic tuning of spatiotemporal simulation models using external data.

  • Embedded Systems - anomaly detection in automative systems and system traces

  • Network Communication Security Breach Detection - analyzing network meta-data to classify anomalous patterns of usage which may indicate security breaches in a communication network. 

  • Invasive Species - treatment of plant species along river banks to inhibit spread of invasive plants using Markov decision processes. 

  • Social Networks - using insights from spatially spreading processes to analyse spread of information and behaviours in social networks


Broad Concepts

Some of the related concepts of interest to the lab's research.

Uncertainty, Causality and Interpretability


Spatially Spreading Processes

Learning, modelling, predicting spatially spreading processes in multiple diverse domains such as Forest Wildfire management, Invasive Species management and Brain Imaging.

Computational Sustainability

The field of Computational Sustainability is at the intersection between computational sciences (such as artificial intelligence, computational modelling, optimization and planning research) with applied research in environmental/ecological domains (such as land use management, invasive species spread, sustainable ecology management, smart grids and species tracking)


In any system with rich states which can change and interact over time there is the possibility that small change in dynamics, initial state or action policy could lead to large changes in the type or value of the outcome. This property is the essence of a complex system and there is a wide variety of research ongoing in different fields in such as social science, engineering, ecology, physics, economics and computer science to understand complex systems and develop ways to bring some order to chaos. Some members of our lab are affiliated with the Waterloo Institute for Complexity and Innovation (WICI) which helps to bring researchers together on these topics.