arXiv subjects

Andrii Krutsylo

Publications and source records attributed to Andrii Krutsylo.

6 recordsLinked to original sources

arXiv2026

Continual Learning via Ensemble-Based Depth-Wise Masked Autoencoders for Data Quality Monitoring in High-Energy Physics

Machine learning (ML) techniques have been demonstrated to improve the accuracy and efficiency of anomaly detection (AD) when compared to conventional methods. This has led to the adoption of ML for data quality monitoring (DQM) use cases in order to monitor the operation of certain systems to ensure that they are free of undesirable or potentially deleterious anomalies. For applications in the field of High-Energy physics (HEP), where detectors must operate in long-running, harsh environments, ML models used in DQM that have been trained on static datasets are bound to experience degraded performance due to distributional shifts that naturally occur in the incoming data streams, unless directly mitigated via the inclusion of continual ML techniques. This work introduces DepthViT, a lightweight masked autoencoder architecture that employs unique depth-wise embeddings and depth-wise attention, to perform computationally efficient AD tasks. A continual learning framework is developed in which DepthViT models trained on the most recent data streams are ensembled with older models to create a robust overall system which is more resilient to shifts in incoming data streams. When evaluated on occupancy maps from the Compact Muon Solenoid (CMS) hadron calorimeter across multiple data-taking campaigns, the proposed method maintains precision above 98.8\% and a stable ratio of correct anomaly predictions to number of anomalies both under small and large distributional shifts. Beyond HEP, the same ensembling-based continual adaptation strategy can be directly applied to industrial monitoring environments where data also naturally evolve over time. This work therefore presents a path toward adaptive anomaly detection systems capable of sustained operation in dynamic data environments.

hep-ex↗

arXiv2025

Non-Uniform Memory Sampling in Experience Replay

Continual learning is the process of training machine learning models on a sequence of tasks where data distributions change over time. A well-known obstacle in this setting is catastrophic forgetting, a phenomenon in which a model drastically loses performance on previously learned tasks when learning new ones. A popular strategy to alleviate this problem is experience replay, in which a subset of old samples is stored in a memory buffer and replayed with new data. Despite continual learning advances focusing on which examples to store and how to incorporate them into the training loss, most approaches assume that sampling from this buffer is uniform by default. We challenge the assumption that uniform sampling is necessarily optimal. We conduct an experiment in which the memory buffer updates the same way in every trial, but the replay probability of each stored sample changes between trials based on different random weight distributions. Specifically, we generate 50 different non-uniform sampling probability weights for each trial and compare their final accuracy to the uniform sampling baseline. We find that there is always at least one distribution that significantly outperforms the baseline across multiple buffer sizes, models, and datasets. These results suggest that more principled adaptive replay policies could yield further gains. We discuss how exploiting this insight could inspire new research on non-uniform memory sampling in continual learning to better mitigate catastrophic forgetting. The code supporting this study is available at $\href{https://github.com/DentonJC/memory-sampling}{https://github.com/DentonJC/memory-sampling}$.

cs.LG↗

arXiv2025

Scalable Forward-Forward Algorithm

We propose a scalable Forward-Forward (FF) algorithm that eliminates the need for backpropagation by training each layer separately. Unlike backpropagation, FF avoids backward gradients and can be more modular and memory efficient, making it appealing for large networks. We extend FF to modern convolutional architectures, such as MobileNetV3 and ResNet18, by introducing a new way to compute losses for convolutional layers. Experiments show that our method achieves performance comparable to standard backpropagation. Furthermore, when we divide the network into blocks, such as the residual blocks in ResNet, and apply backpropagation only within each block, but not across blocks, our hybrid design tends to outperform backpropagation baselines while maintaining a similar training speed. Finally, we present experiments on small datasets and transfer learning that confirm the adaptability of our method.

cs.LG↗

arXiv2024

Continually Learn to Map Visual Concepts to Large Language Models in Resource-constrained Environments

Learning continually from a stream of non-i.i.d. data is an open challenge in deep learning, even more so when working in resource-constrained environments such as embedded devices. Visual models that are continually updated through supervised learning are often prone to overfitting, catastrophic forgetting, and biased representations. On the other hand, large language models contain knowledge about multiple concepts and their relations, which can foster a more robust, informed and coherent learning process. This work proposes Continual Visual Mapping (CVM), an approach that continually ground vision representations to a knowledge space extracted from a fixed Language model. Specifically, CVM continually trains a small and efficient visual model to map its representations into a conceptual space established by a fixed Large Language Model. Due to their smaller nature, CVM can be used when directly adapting large visual pre-trained models is unfeasible due to computational or data constraints. CVM overcome state-of-the-art continual learning methods on five benchmarks and offers a promising avenue for addressing generalization capabilities in continual learning, even in computationally constrained devices.

cs.AI↗

arXiv2023

DIT4BEARs Smart Roads Internship

The research internship at UiT - The Arctic University of Norway was offered for our team being the winner of the 'Smart Roads - Winter Road Maintenance 2021' Hackathon. The internship commenced on 3 May 2021 and ended on 21 May 2021 with meetings happening twice each week. In spite of having different nationalities and educational backgrounds, we both interns tried to collaborate as a team as much as possible. The most alluring part was working on this project made us realize the critical conditions faced by the arctic people, where it was hard to gain such a unique experience from our residence. We developed and implemented several deep learning models to classify the states (dry, moist, wet, icy, snowy, slushy). Depending upon the best model, the weather forecast app will predict the state taking the Ta, Tsurf, Height, Speed, Water, etc. into consideration. The crucial part was to define a safety metric which is the product of the accident rates based on friction and the accident rates based on states. We developed a regressor that will predict the safety metric depending upon the state obtained from the classifier and the friction obtained from the sensor data. A pathfinding algorithm has been designed using the sensor data, open street map data, weather data.

cs.LG↗

arXiv2022

Hebbian Continual Representation Learning

Continual Learning aims to bring machine learning into a more realistic scenario, where tasks are learned sequentially and the i.i.d. assumption is not preserved. Although this setting is natural for biological systems, it proves very difficult for machine learning models such as artificial neural networks. To reduce this performance gap, we investigate the question whether biologically inspired Hebbian learning is useful for tackling continual challenges. In particular, we highlight a realistic and often overlooked unsupervised setting, where the learner has to build representations without any supervision. By combining sparse neural networks with Hebbian learning principle, we build a simple yet effective alternative (HebbCL) to typical neural network models trained via the gradient descent. Due to Hebbian learning, the network have easily interpretable weights, which might be essential in critical application such as security or healthcare. We demonstrate the efficacy of HebbCL in an unsupervised learning setting applied to MNIST and Omniglot datasets. We also adapt the algorithm to the supervised scenario and obtain promising results in the class-incremental learning.

cs.NE↗