Eugene Belilovsky

Paul Janson

Doctorat - Concordia

Co-superviseur⋅e :

Charles-Etienne Joseph

Maîtrise recherche - UdeM

Co-superviseur⋅e :

Zafir Khalid

Maîtrise recherche - Concordia

Co-superviseur⋅e :

Irina Rish

Site web

Gwen Legate

Doctorat - Concordia

Co-superviseur⋅e :

Maîtrise recherche - Concordia

Co-superviseur⋅e :

Melika Minaei Bidgoli

Stagiaire de recherche - Concordia University

Doctorat - Concordia

Adel Nabli

Doctorat - Concordia

Google Scholar

Geraldin Nanfack

Postdoctorat - Concordia

Co-superviseur⋅e :

geraldin.nanfack@mila.quebec

Site web

Google Scholar

Albert Orozco Camacho

Doctorat - Concordia

Co-superviseur⋅e :

Doctorat - Concordia

Co-superviseur⋅e :

Irina Rish

Benjamin Therien

Doctorat - UdeM

Superviseur⋅e principal⋅e :

Collaborateur·rice de recherche - UdeM

Superviseur⋅e principal⋅e :

Maîtrise recherche - Concordia

AmirHossein Zamani

Doctorat - Concordia

Co-superviseur⋅e :

Congshu Zou

Maîtrise recherche - Concordia

Publications

WASH: Train your Ensemble with Communication-Efficient Weight Shuffling, then Average

Louis Fournier

Adel Nabli

Masih Aminbeidokhti

Marco Pedersoli

Edouard Oyallon

The performance of deep neural networks is enhanced by ensemble methods, which average the output of several models. However, this comes at … (voir plus)an increased cost at inference. Weight averaging methods aim at balancing the generalization of ensembling and the inference speed of a single model by averaging the parameters of an ensemble of models. Yet, naive averaging results in poor performance as models converge to different loss basins, and aligning the models to improve the performance of the average is challenging. Alternatively, inspired by distributed training, methods like DART and PAPA have been proposed to train several models in parallel such that they will end up in the same basin, resulting in good averaging accuracy. However, these methods either compromise ensembling accuracy or demand significant communication between models during training. In this paper, we introduce WASH, a novel distributed method for training model ensembles for weight averaging that achieves state-of-the-art image classification accuracy. WASH maintains models within the same basin by randomly shuffling a small percentage of weights during training, resulting in diverse models and lower communication costs compared to standard parameter averaging methods.

2024-10-10

NeurIPS.cc/2024/Workshop/OPT (publié)

Not Only the Last-Layer Features for Spurious Correlations: All Layer Deep Feature Reweighting

Humza Wajid Hameed

G'eraldin Nanfack

Spurious correlations are a major source of errors for machine learning models, in particular when aiming for group-level fairness. It has b… (voir plus)een recently shown that a powerful approach to combat spurious correlations is to re-train the last layer on a balanced validation dataset, isolating robust features for the predictor. However, key attributes can sometimes be discarded by neural networks towards the last layer. In this work, we thus consider retraining a classifier on a set of features derived from all layers. We utilize a recently proposed feature selection strategy to select unbiased features from all the layers. We observe this approach gives significant improvements in worst-group accuracy on several standard benchmarks.

2024-09-23

ArXiv (prépublication)

Accelerating Training with Neuron Interaction and Nowcasting Networks

Boris Knyazev

Abhinav Moudgil

Guillaume Lajoie

Simon Lacoste-Julien

Neural network training can be accelerated when a learnable update rule is used in lieu of classic adaptive optimizers (e.g. Adam). However,… (voir plus) learnable update rules can be costly and unstable to train and use. Recently, Jang et al. (2023) proposed a simpler approach to accelerate training based on weight nowcaster networks (WNNs). In their approach, Adam is used for most of the optimization steps and periodically, only every few steps, a WNN nowcasts (predicts near future) parameters. We improve WNNs by proposing neuron interaction and nowcasting (NiNo) networks. In contrast to WNNs, NiNo leverages neuron connectivity and graph neural networks to more accurately nowcast parameters. We further show that in some networks, such as Transformers, modeling neuron connectivity accurately is challenging. We address this and other limitations, which allows NiNo to accelerate Adam training by up to 50% in vision and language tasks.

2024-09-06

ArXiv (prépublication)

Harmony in Diversity: Merging Neural Networks with Canonical Correlation Analysis

Stefan Horoi

Albert Manuel Orozco Camacho

2024-07-08

Proceedings of the 41st International Conference on Machine Learning (publié)

Simple and Scalable Strategies to Continually Pre-train Large Language Models

Adam Ibrahim

Benjamin Thérien

Kshitij Gupta

Mats Leon Richter

Quentin Gregory Anthony

Timothee LESORT

Irina Rish

2024-07-08

TMLR (accepté)

Harmony in Diversity: Merging Neural Networks with Canonical Correlation Analysis

Stefan Horoi

Albert Manuel Orozco Camacho

Combining the predictions of multiple trained models through ensembling is generally a good way to improve accuracy by leveraging the differ… (voir plus)ent learned features of the models, however it comes with high computational and storage costs. Model fusion, the act of merging multiple models into one by combining their parameters reduces these costs but doesn't work as well in practice. Indeed, neural network loss landscapes are high-dimensional and non-convex and the minima found through learning are typically separated by high loss barriers. Numerous recent works have been focused on finding permutations matching one network features to the features of a second one, lowering the loss barrier on the linear path between them in parameter space. However, permutations are restrictive since they assume a one-to-one mapping between the different models' neurons exists. We propose a new model merging algorithm, CCA Merge, which is based on Canonical Correlation Analysis and aims to maximize the correlations between linear combinations of the model features. We show that our alignment method leads to better performances than past methods when averaging models trained on the same, or differing data splits. We also extend this analysis into the harder setting where more than 2 models are merged, and we find that CCA Merge works significantly better than past methods. Our code is publicly available at https://github.com/shoroi/align-n-merge

2024-07-07

ArXiv (prépublication)

Harmony in Diversity: Merging Neural Networks with Canonical Correlation Analysis

Stefan Horoi

Albert Manuel Orozco Camacho

2024-07-07

ArXiv (prépublication)

Model Breadcrumbs: Scalable Upcycling of Finetuned Foundation Models via Sparse Task Vectors Merging

MohammadReza Davari

2024-07-03

ICML.cc/2024/Workshop/FM-Wild (poster)

Simulating federated learning for steatosis detection using ultrasound images

Yue Qi

Pedro Vianna

Alexandre Cadrin-Chênevert

Katleen Blanchet

Emmanuel Montagnon

Louis-Antoine Mullie

Guy Cloutier

Michael Chassé

An Tang

2024-06-10

Scientific Reports (publié)

PETRA: Parallel End-to-end Training with Reversible Architectures

Stephane Rivaud

Louis Fournier

Thomas Pumir

Michael Eickenberg

Edouard Oyallon

Reversible architectures have been shown to be capable of performing on par with their non-reversible architectures, being applied in deep l… (voir plus)earning for memory savings and generative modeling. In this work, we show how reversible architectures can solve challenges in parallelizing deep model training. We introduce PETRA, a novel alternative to backpropagation for parallelizing gradient computations. PETRA facilitates effective model parallelism by enabling stages (i.e., a set of layers) to compute independently on different devices, while only needing to communicate activations and gradients between each other. By decoupling the forward and backward passes and keeping a single updated version of the parameters, the need for weight stashing is also removed. We develop a custom autograd-like training framework for PETRA, and we demonstrate its effectiveness on CIFAR-10, ImageNet32, and ImageNet, achieving competitive accuracies comparable to backpropagation using ResNet-18, ResNet-34, and ResNet-50 models.

2024-06-04

ArXiv (prépublication)

ACCO: Accumulate while you Communicate, Hiding Communications in Distributed LLM Training

Adel Nabli

Louis Fournier

Pierre ERBACHER

Louis Serrano

Edouard Oyallon

Training Large Language Models (LLMs) relies heavily on distributed implementations, employing multiple GPUs to compute stochastic gradients… (voir plus) on model replicas in parallel. However, synchronizing gradients in data parallel settings induces a communication overhead increasing with the number of distributed workers, which can impede the efficiency gains of parallelization. To address this challenge, optimization algorithms reducing inter-worker communication have emerged, such as local optimization methods used in Federated Learning. While effective in minimizing communication overhead, these methods incur significant memory costs, hindering scalability: in addition to extra momentum variables, if communications are only allowed between multiple local optimization steps, then the optimizer's states cannot be sharded among workers. In response, we propose

2024-06-03

ArXiv (prépublication)

From Feature Visualization to Visual Circuits: Effect of Adversarial Model Manipulation

G'eraldin Nanfack

Michael Eickenberg

Understanding the inner working functionality of large-scale deep neural networks is challenging yet crucial in several high-stakes applicat… (voir plus)ions. Mechanistic inter- pretability is an emergent field that tackles this challenge, often by identifying human-understandable subgraphs in deep neural networks known as circuits. In vision-pretrained models, these subgraphs are usually interpreted by visualizing their node features through a popular technique called feature visualization. Recent works have analyzed the stability of different feature visualization types under the adversarial model manipulation framework. This paper starts by addressing limitations in existing works by proposing a novel attack called ProxPulse that simultaneously manipulates the two types of feature visualizations. Surprisingly, when analyzing these attacks under the umbrella of visual circuits, we find that visual circuits show some robustness to ProxPulse. We, therefore, introduce a new attack based on ProxPulse that unveils the manipulability of visual circuits, shedding light on their lack of robustness. The effectiveness of these attacks is validated using pre-trained AlexNet and ResNet-50 models on ImageNet.

2024-06-03

ArXiv (prépublication)