Abdul Wahab

wahab1 (at) ualberta (dot) ca

I’m a Machine Learning Scientist at Amii, where I lead applied AI projects involving reinforcement learning, large language models, and multi-agent systems across domains such as smart grids, robotics, field service, games, and healthcare.

Previously, I was an Associate Researcher at Huawei Canada, where I worked on personalized reinforcement learning agents, image quality assessment, and the camera ISP pipeline.

I completed my master’s in computer science at the University of Alberta with Martha White. My research focused on exploration in reinforcement learning.

CV  /  LinkedIn  /  Google Scholar  /  Github

We propose a new task called Distortion Graph (DG), which represents pairs of images as a region-grounded topology. The graph compactly captures dense degradation information, including distortion type, severity, pairwise comparison, and quality score, in an interpretable structure.

Reward bonus methods for exploration do not provide first-visit optimism because they propagate local uncertainty only after it has been observed. We extend reward bonuses to value bonuses and propose Value Bonuses with Ensemble Errors (VBE), which maintains an ensemble of random action-value functions (RQFs) and uses their estimation errors to provide first-visit optimism and deep exploration.

We propose a general method based on bi-directional sequential models (LSTMs) for short-term energy load forecasting suitable for both individual household energy forecasting and regional energy forecasting.

We propose and compare algorithms to allocate upstream PON capacity, where multiple virtual operators generate independent frame-level allocation over a shared infrastructure. Our fragmentation-aware approach shows the ability to limit latency increase to a few microseconds.

We use Two-timescale networks (TTN) in an offline-online setting: an agent is first trained on offline data, then updates its representations and policy online. We empirically study sparsity (FTA), input transformations, data augmentation, and self-supervised contrastive objectives to identify which methods contribute most to effective representations and stable online learning.

In this project, we couple combinatorial game theory (CGT) enhancements such as an endgame database and subgame simplifications with the Monte Carlo Tree Search (MCTS) solver to design a strong 2D Clobber player.

Human visual system depends on imagery representations in addition to the visual input to build a concrete holistic picture of the object of interest. These representations are correlated and shared in the human brain and help provide a crisp understanding of the environment. In this work, we study if a similar correlation exists in Artificial Neural Networks (ANNs) and if the correlation is strong enough to use the corresponding fMRI data of both perceptual and quasi-perceptual experiences interchangeably for downstream tasks such as object category prediction.

In this work, we explored the advantages of using a line-scan camera in industrial applications, where the point of observation is a single plane, for monitoring, classification, and segmentation of moving objects. A ConvLSTM-based architecture was proposed, which extracts temporal-correlation from the sequence of line-scans to compensate for the loss of spatial-information in the individual line-scan (single row of pixels). The methodology evaluated on road-crack segmentation datasets outperforms the current benchmark on four public datasets. The methodology was also evaluated on a novel dataset collected on the roads of Islamabad, Pakistan. The methodology was also implemented for a seed-sorting system.

Abdul Wahab. Design inspired by this website.