Yani Ioannou

Mohammed Adnan, Rohan Jain, Tom Jacobs et al.

Dynamic Sparse Training (DST) methods train neural networks by maintaining sparsity while dynamically adapting the network topology. Despite the promise of reduced computation, DST methods converge significantly slower than dense training, often requiring comparable training time to achieve similar accuracy. We demonstrate both analytically and empirically that Batch Normalization (BN) adversely affects sparse training, and propose SparseOpt, a sparsity-aware optimizer, to address this. Experiments on ResNet models across CIFAR-100 and ImageNet demonstrate consistently faster convergence and improved generalization with our proposed method. Our work highlights the limitations of current normalization layers in sparse training and provides the first systematic study of the interaction between Batch Normalization, sparse layers, and DST, taking a significant step toward making DST practically competitive with dense training.

Mohammed Adnan, Yani Ioannou, Chuan-Yung Tsai et al.

The failure of deep neural networks to generalize to out-of-distribution data is a well-known problem and raises concerns about the deployment of trained networks in safety-critical domains such as healthcare, finance and autonomous vehicles. We study a particular kind of distribution shift $\unicode{x2013}$ shortcuts or spurious correlations in the training data. Shortcut learning is often only exposed when models are evaluated on real-world data that does not contain the same spurious correlations, posing a serious dilemma for AI practitioners to properly assess the effectiveness of a trained model for real-world applications. In this work, we propose to use the mutual information (MI) between the learned representation and the input as a metric to find where in training, the network latches onto shortcuts. Experiments demonstrate that MI can be used as a domain-agnostic metric for monitoring shortcut learning.

Angus Galloway, Anna Golubeva, Mahmoud Salem et al.

Estimating the Generalization Error (GE) of Deep Neural Networks (DNNs) is an important task that often relies on availability of held-out data. The ability to better predict GE based on a single training set may yield overarching DNN design principles to reduce a reliance on trial-and-error, along with other performance assessment advantages. In search of a quantity relevant to GE, we investigate the Mutual Information (MI) between the input and final layer representations, using the infinite-width DNN limit to bound MI. An existing input compression-based GE bound is used to link MI and GE. To the best of our knowledge, this represents the first empirical study of this bound. In our attempt to empirically falsify the theoretical bound, we find that it is often tight for best-performing models. Furthermore, it detects randomization of training labels in many cases, reflects test-time perturbation robustness, and works well given only few training samples. These results are promising given that input compression is broadly applicable where MI can be estimated with confidence.

Farzaneh Dehghani, Mahsa Dibaji, Fahim Anzum et al.

Artificial Intelligence (AI) has paved the way for revolutionary decision-making processes, which if harnessed appropriately, can contribute to advancements in various sectors, from healthcare to economics. However, its black box nature presents significant ethical challenges related to bias and transparency. AI applications are hugely impacted by biases, presenting inconsistent and unreliable findings, leading to significant costs and consequences, highlighting and perpetuating inequalities and unequal access to resources. Hence, developing safe, reliable, ethical, and Trustworthy AI systems is essential. Our team of researchers working with Trustworthy and Responsible AI, part of the Transdisciplinary Scholarship Initiative within the University of Calgary, conducts research on Trustworthy and Responsible AI, including fairness, bias mitigation, reproducibility, generalization, interpretability, and authenticity. In this paper, we review and discuss the intricacies of AI biases, definitions, methods of detection and mitigation, and metrics for evaluating bias. We also discuss open challenges with regard to the trustworthiness and widespread application of AI across diverse domains of human-centric decision making, as well as guidelines to foster Responsible and Trustworthy AI models.

Sindhuja Chaduvula, Ahmed Y. Radwan, Azib Farooq et al.

Preference alignment methods such as RLHF and Direct Preference Optimization (DPO) improve instruction following, but they can also reinforce hallucinations when preference judgments reward fluency and confidence over factual correctness. We introduce F-DPO (Factuality-aware Direct Preference Optimization), a simple extension of DPO that uses only binary factuality labels. F-DPO (i) applies a label-flipping transformation that corrects misordered preference pairs so the chosen response is never less factual than the rejected one, and (ii) adds a factuality-aware margin that emphasizes pairs with clear correctness differences, while reducing to standard DPO when both responses share the same factuality. We construct factuality-aware preference data by augmenting DPO pairs with binary factuality indicators and synthetic hallucinated variants. Across seven open-weight LLMs (1B-14B), F-DPO consistently improves factuality and reduces hallucination rates relative to both base models and standard DPO. On Qwen3-8B, F-DPO reduces hallucination rates by 5x(from 0.424 to 0.084) while improving factuality scores by 50% (from 5.26 to 7.90). F-DPO also generalizes to out-of-distribution benchmarks: on TruthfulQA, Qwen2.5-14B achieves +17% MC1 accuracy (0.500 to 0.585) and +49% MC2 accuracy (0.357 to 0.531). F-DPO requires no auxiliary reward model, token-level annotations, or multi-stage training.

Cem Üyük, Mike Lasby, Mohamed Yassin et al.

Large neural networks exhibit exceptional performance across numerous tasks, yet their considerable size often hinders deployment on resource-constrained systems. While various model compression strategies have been well studied, parameter sharing remains underexplored. In this paper, we introduce Fine-grained Parameter Sharing (FiPS), a novel algorithm that leverages parameter sharing, tensor decomposition, and sparsity to effectively compress large-scale Vision Transformers (ViTs) and Large Language Models (LLMs). FiPS employs a shared base and sparse factors to represent neurons across multi-layer perceptron (MLP) modules, where initialization is guided by Singular Value Decomposition (SVD) and subsequent optimization is conducted through block-wise reconstruction error minimization. Experimental results show that FiPS reduces the parameter budget of MLP modules by 50-75% for DeiT-B and Swin-L and by 40-50% for various Gemma-2 and Llama-3 models while maintaining ViT model accuracy within 1% pt. of the original and LLM perplexity with negligible degradation.

Nasib Ullah, Erik Schultheis, Mike Lasby et al.

In recent years, Dynamic Sparse Training (DST) has emerged as an alternative to post-training pruning for generating efficient models. In principle, DST allows for a more memory efficient training process, as it maintains sparsity throughout the entire training run. However, current DST implementations fail to capitalize on this in practice. Because sparse matrix multiplication is much less efficient than dense matrix multiplication on GPUs, most implementations simulate sparsity by masking weights. In this paper, we leverage recent advances in semi-structured sparse training to apply DST in the domain of classification with large output spaces, where memory-efficiency is paramount. With a label space of possibly millions of candidates, the classification layer alone will consume several gigabytes of memory. Switching from a dense to a fixed fan-in sparse layer updated with sparse evolutionary training (SET); however, severely hampers training convergence, especially at the largest label spaces. We find that poor gradient flow from the sparse classifier to the dense text encoder make it difficult to learn good input representations. By employing an intermediate layer or adding an auxiliary training objective, we recover most of the generalisation performance of the dense model. Overall, we demonstrate the applicability and practical benefits of DST in a challenging domain -- characterized by a highly skewed label distribution that differs substantially from typical DST benchmark datasets -- which enables end-to-end training with millions of labels on commodity hardware.

Mohammed Adnan, Rohan Jain, Ekansh Sharma et al.

The Lottery Ticket Hypothesis (LTH) suggests there exists a sparse LTH mask and weights that achieve the same generalization performance as the dense model while using significantly fewer parameters. However, finding a LTH solution is computationally expensive, and a LTH sparsity mask does not generalize to other random weight initializations. Recent work has suggested that neural networks trained from random initialization find solutions within the same basin modulo permutation, and proposes a method to align trained models within the same loss basin. We hypothesize that misalignment of basins is the reason why LTH masks do not generalize to new random initializations and propose permuting the LTH mask to align with the new optimization basin when performing sparse training from a different random init. We empirically show a significant increase in generalization when sparse training from random initialization with the permuted mask as compared to using the non-permuted LTH mask, on multiple datasets (CIFAR-10, CIFAR-100 and ImageNet) and models (VGG11, ResNet20 and ResNet50).

Mike Lasby, Nish Sinnadurai, Valavan Manohararajah et al.

Speculative decoding is a powerful technique for reducing the latency of Large Language Models (LLMs), offering a fault-tolerant framework that enables the use of highly compressed draft models. In this work, we introduce Self-Distilled Sparse Drafters (SD$^2$), a novel methodology that leverages self-data distillation and fine-grained weight sparsity to produce highly efficient and well-aligned draft models. SD$^2$ systematically enhances draft token acceptance rates while significantly reducing Multiply-Accumulate operations (MACs), even in the Universal Assisted Generation (UAG) setting, where draft and target models originate from different model families. On a Llama-3.1-70B target model, SD$^2$ provides a 1.59$\times$ higher Mean Accepted Length (MAL) compared to layer-pruned draft models and reduces MACs by over 43.87% with a 8.36% reduction in MAL compared to a dense draft models. Our 1.5B and 3B unstructured sparse drafters outperform both dense and layer-pruned models in terms of end-to-end latency improvements; highlighting the potential of sparsity-aware fine-tuning and compression strategies to improve LLM inference efficiency while maintaining alignment with target models.

Mike Lasby, Ivan Lazarevich, Nish Sinnadurai et al.

Sparsely-activated Mixture-of-Experts (SMoE) models offer efficient pre-training and low latency but their large parameter counts create significant memory overhead, motivating research into expert compression. Contrary to recent findings favouring expert merging on discriminative benchmarks, we demonstrate that expert pruning is a superior strategy for generative tasks. We prove that merging introduces an irreducible error by causing a "functional subspace collapse", due to the loss of the router's independent, input-dependent control over experts. Leveraging this insight, we propose Router-weighted Expert Activation Pruning (REAP), a novel pruning criterion that considers both router gate-values and expert activation norms. Across a diverse set of SMoE models ranging from 20B to 1T parameters, REAP consistently outperforms merging and other pruning methods on generative benchmarks, especially at 50% compression. Notably, our method achieves near-lossless compression on code generation and tool-calling tasks with Qwen3-Coder-480B and Kimi-K2, even after pruning 50% of experts.

Mahdi Mohammadizadeh, Arash Mozhdehi, Yani Ioannou et al.

Although many real-world applications, such as disease prediction, and fault detection suffer from class imbalance, most existing graph-based classification methods ignore the skewness of the distribution of classes; therefore, tend to be biased towards the majority class(es). Conventional methods typically tackle this problem through the assignment of weights to each one of the class samples based on a function of their loss, which can lead to over-fitting on outliers. In this paper, we propose a meta-learning algorithm, named Meta-GCN, for adaptively learning the example weights by simultaneously minimizing the unbiased meta-data set loss and optimizing the model weights through the use of a small unbiased meta-data set. Through experiments, we have shown that Meta-GCN outperforms state-of-the-art frameworks and other baselines in terms of accuracy, the area under the receiver operating characteristic (AUC-ROC) curve, and macro F1-Score for classification tasks on two different datasets.

Mike Lasby, Anna Golubeva, Utku Evci et al.

Dynamic Sparse Training (DST) methods achieve state-of-the-art results in sparse neural network training, matching the generalization of dense models while enabling sparse training and inference. Although the resulting models are highly sparse and theoretically less computationally expensive, achieving speedups with unstructured sparsity on real-world hardware is challenging. In this work, we propose a sparse-to-sparse DST method, Structured RigL (SRigL), to learn a variant of fine-grained structured N:M sparsity by imposing a constant fan-in constraint. Using our empirical analysis of existing DST methods at high sparsity, we additionally employ a neuron ablation method which enables SRigL to achieve state-of-the-art sparse-to-sparse structured DST performance on a variety of Neural Network (NN) architectures. Using a 90% sparse linear layer, we demonstrate a real-world acceleration of 3.4x/2.5x on CPU for online inference and 1.7x/13.0x on GPU for inference with a batch size of 256 when compared to equivalent dense/unstructured (CSR) sparse layers, respectively.

Osbert Bastani, Yani Ioannou, Leonidas Lampropoulos et al.

Despite having high accuracy, neural nets have been shown to be susceptible to adversarial examples, where a small perturbation to an input can cause it to become mislabeled. We propose metrics for measuring the robustness of a neural net and devise a novel algorithm for approximating these metrics based on an encoding of robustness as a linear program. We show how our metrics can be used to evaluate the robustness of deep neural nets with experiments on the MNIST and CIFAR-10 datasets. Our algorithm generates more informative estimates of robustness metrics compared to estimates based on existing algorithms. Furthermore, we show how existing approaches to improving robustness "overfit" to adversarial examples generated using a specific algorithm. Finally, we show that our techniques can be used to additionally improve neural net robustness both according to the metrics that we propose, but also according to previously proposed metrics.

Yani Ioannou, Duncan Robertson, Roberto Cipolla et al.

We propose a new method for creating computationally efficient and compact convolutional neural networks (CNNs) using a novel sparse connection structure that resembles a tree root. This allows a significant reduction in computational cost and number of parameters compared to state-of-the-art deep CNNs, without compromising accuracy, by exploiting the sparsity of inter-layer filter dependencies. We validate our approach by using it to train more efficient variants of state-of-the-art CNN architectures, evaluated on the CIFAR10 and ILSVRC datasets. Our results show similar or higher accuracy than the baseline architectures with much less computation, as measured by CPU and GPU timings. For example, for ResNet 50, our model has 40% fewer parameters, 45% fewer floating point operations, and is 31% (12%) faster on a CPU (GPU). For the deeper ResNet 200 our model has 25% fewer floating point operations and 44% fewer parameters, while maintaining state-of-the-art accuracy. For GoogLeNet, our model has 7% fewer parameters and is 21% (16%) faster on a CPU (GPU).

Sukrit Shankar, Duncan Robertson, Yani Ioannou et al.

Deep Convolutional Neural Networks (CNNs) have recently evinced immense success for various image recognition tasks. However, a question of paramount importance is somewhat unanswered in deep learning research - is the selected CNN optimal for the dataset in terms of accuracy and model size? In this paper, we intend to answer this question and introduce a novel strategy that alters the architecture of a given CNN for a specified dataset, to potentially enhance the original accuracy while possibly reducing the model size. We use two operations for architecture refinement, viz. stretching and symmetrical splitting. Our procedure starts with a pre-trained CNN for a given dataset, and optimally decides the stretch and split factors across the network to refine the architecture. We empirically demonstrate the necessity of the two operations. We evaluate our approach on two natural scenes attributes datasets, SUN Attributes and CAMIT-NSAD, with architectures of GoogleNet and VGG-11, that are quite contrasting in their construction. We justify our choice of datasets, and show that they are interestingly distinct from each other, and together pose a challenge to our architectural refinement algorithm. Our results substantiate the usefulness of the proposed method.

Yani Ioannou, Duncan Robertson, Darko Zikic et al.

This paper investigates the connections between two state of the art classifiers: decision forests (DFs, including decision jungles) and convolutional neural networks (CNNs). Decision forests are computationally efficient thanks to their conditional computation property (computation is confined to only a small region of the tree, the nodes along a single branch). CNNs achieve state of the art accuracy, thanks to their representation learning capabilities. We present a systematic analysis of how to fuse conditional computation with representation learning and achieve a continuum of hybrid models with different ratios of accuracy vs. efficiency. We call this new family of hybrid models conditional networks. Conditional networks can be thought of as: i) decision trees augmented with data transformation operators, or ii) CNNs, with block-diagonal sparse weight matrices, and explicit data routing functions. Experimental validation is performed on the common task of image classification on both the CIFAR and Imagenet datasets. Compared to state of the art CNNs, our hybrid models yield the same accuracy with a fraction of the compute cost and much smaller number of parameters.

Yani Ioannou, Duncan Robertson, Jamie Shotton et al.

We propose a new method for creating computationally efficient convolutional neural networks (CNNs) by using low-rank representations of convolutional filters. Rather than approximating filters in previously-trained networks with more efficient versions, we learn a set of small basis filters from scratch; during training, the network learns to combine these basis filters into more complex filters that are discriminative for image classification. To train such networks, a novel weight initialization scheme is used. This allows effective initialization of connection weights in convolutional layers composed of groups of differently-shaped filters. We validate our approach by applying it to several existing CNN architectures and training these networks from scratch using the CIFAR, ILSVRC and MIT Places datasets. Our results show similar or higher accuracy than conventional CNNs with much less compute. Applying our method to an improved version of VGG-11 network using global max-pooling, we achieve comparable validation accuracy using 41% less compute and only 24% of the original VGG-11 model parameters; another variant of our method gives a 1 percentage point increase in accuracy over our improved VGG-11 model, giving a top-5 center-crop validation accuracy of 89.7% while reducing computation by 16% relative to the original VGG-11 model. Applying our method to the GoogLeNet architecture for ILSVRC, we achieved comparable accuracy with 26% less compute and 41% fewer model parameters. Applying our method to a near state-of-the-art network for CIFAR, we achieved comparable accuracy with 46% less compute and 55% fewer parameters.

Yani Ioannou, Babak Taati, Robin Harrap et al.

A novel multi-scale operator for unorganized 3D point clouds is introduced. The Difference of Normals (DoN) provides a computationally efficient, multi-scale approach to processing large unorganized 3D point clouds. The application of DoN in the multi-scale filtering of two different real-world outdoor urban LIDAR scene datasets is quantitatively and qualitatively demonstrated. In both datasets the DoN operator is shown to segment large 3D point clouds into scale-salient clusters, such as cars, people, and lamp posts towards applications in semi-automatic annotation, and as a pre-processing step in automatic object recognition. The application of the operator to segmentation is evaluated on a large public dataset of outdoor LIDAR scenes with ground truth annotations.