DM
Decebal Constantin Mocanu
cs.LGcs.AIcs.CVcs.NEstat.MLAlgorithmscontinuous learningcs.CLcs.MAcs.MM
On Valency
published · living versionsW_g237bs4b·v1 · currentpublished
Dr. No: Feature Steering for Abstention in Vision-Language Models
with Farid Hasanov, Riccardo Renzulli, Bahram Zonooz, Mykola Pechenizkiy
1 version
W_wy892x7w·v1published
Preprints & journals
57 papers in the corpus · 2016–2026When Data Is Scarce: Scaling Sparse Language Models with Repeated Training2606.01155v1 · Boqian Wu, Qiao Xiao, Patrik Okanovic et al.2026 · 0 citationsarXiv
Memory-Efficient LLM Training with Dynamic Sparsity: From Stability to Practical Scaling2606.00888v1 · Qiao Xiao, Boqian Wu, Patrik Okanovic et al.2026 · 0 citationsarXiv
Batch Matrix-form Equations and Implementation of Multilayer Perceptrons2511.11918v1 · Wieger Wesselink, Bram Grooten, Huub van de Wetering et al.2025 · 0 citationsarXiv
Addressing the Collaboration Dilemma in Low-Data Federated Learning via Transient Sparsity2506.00932v1 · Qiao Xiao, Boqian Wu, Andrey Poddubnyy et al.2025 · 0 citationsarXiv
Leave it to the Specialist: Repair Sparse LLMs with Sparse Fine-Tuning via Sparsity Evolution2505.24037v3 · Qiao Xiao, Alan Ansell, Boqian Wu et al.2025 · 0 citationsarXiv
NeuroTrails: Training with Dynamic Sparse Heads as the Key to Effective Ensembling2505.17909v1 · Bram Grooten, Farid Hasanov, Chenxiang Zhang et al.2025 · 0 citationsarXiv
Sparse-to-Sparse Training of Diffusion Models2504.21380v2 · Ines Cardoso Oliveira, Decebal Constantin Mocanu, Luis A. Leiva2025 · 0 citationsTransactions on Machine Learning Research (TMLR) 2025
LiMTR: Time Series Motion Prediction for Diverse Road Users through Multimodal Feature Integration2410.15819v1 · Camiel Oerlemans, Bram Grooten, Michiel Braat et al.2024 · 0 citationsarXiv
Dynamic Sparse Training versus Dense Training: The Unexpected Winner in Image Corruption Robustness2410.03030v2 · Boqian Wu, Qiao Xiao, Shunxin Wang et al.2024 · 0 citationsarXiv
Are Sparse Neural Networks Better Hard Sample Learners?2409.09196v2 · Qiao Xiao, Boqian Wu, Lu Yin et al.2024 · 0 citationsarXiv
Unveiling the Power of Sparse Neural Networks for Feature Selection2408.04583v1 · Zahra Atashgahi, Tennison Liu, Mykola Pechenizkiy et al.2024 · 6 citationsarXiv
Dynamic Data Pruning for Automatic Speech Recognition2406.18373v1 · Qiao Xiao, Pingchuan Ma, Adriana Fernandez-Lopez et al.2024 · 1 citationarXiv
Boosting Robustness in Preference-Based Reinforcement Learning with Dynamic Sparsity2406.06495v3 · Calarina Muslimani, Bram Grooten, Deepak Ranganatha Sastry Mamillapalli et al.2024 · 0 citationsarXiv
Self-Regulated Neurogenesis for Online Data-Incremental Learning2403.14684v2 · Murat Onur Yildirim, Elif Ceren Gok Yildirim, Decebal Constantin Mocanu et al.2024 · 0 citationsarXiv
MaDi: Learning to Mask Distractions for Generalization in Visual Deep Reinforcement Learning2312.15339v1 · Bram Grooten, Tristan Tomilin, Gautham Vasan et al.2023 · 5 citationsarXiv
E2ENet: Dynamic Sparse Feature Fusion for Accurate and Efficient 3D Medical Image Segmentation2312.04727v2 · Boqian Wu, Qiao Xiao, Shiwei Liu et al.2023 · 8 citationsarXiv
Continual Learning with Dynamic Sparse Training: Exploring Algorithms for Effective Model Updates2308.14831v2 · Murat Onur Yildirim, Elif Ceren Gok Yildirim, Ghada Sokar et al.2023 · 9 citationsarXiv
Fantastic Weights and How to Find Them: Where to Prune in Dynamic Sparse Training2306.12230v2 · Aleksandra I. Nowak, Bram Grooten, Decebal Constantin Mocanu et al.2023 · 2 citationsarXiv
Adaptive Sparsity Level during Training for Efficient Time Series Forecasting with Transformers2305.18382v2 · Zahra Atashgahi, Mykola Pechenizkiy, Raymond Veldhuis et al.2023 · 4 citationsarXiv
Supervised Feature Selection with Neuron Evolution in Sparse Neural Networks2303.07200v2 · Zahra Atashgahi, Xuhao Zhang, Neil Kichler et al.2023 · 6 citationsarXiv
Automatic Noise Filtering with Dynamic Sparse Training in Deep Reinforcement Learning2302.06548v1 · Bram Grooten, Ghada Sokar, Shibhansh Dohare et al.2023 · 5 citationsarXiv
Dynamic Sparse Network for Time Series Classification: Learning What to "see''2212.09840v1 · Qiao Xiao, Boqian Wu, Yu Zhang et al.2022 · 22 citationsarXiv
You Can Have Better Graph Neural Networks by Not Training Weights at All: Finding Untrained GNNs Tickets2211.15335v5 · Tianjin Huang, Tianlong Chen, Meng Fang et al.2022 · 3 citationsLoG 2022 (Oral & Best Paper Award)
Where to Pay Attention in Sparse Training for Feature Selection?2211.14627v1 · Ghada Sokar, Zahra Atashgahi, Mykola Pechenizkiy et al.2022 · 7 citationsarXiv
Memory-free Online Change-point Detection: A Novel Neural Network Approach2207.03932v2 · Zahra Atashgahi, Decebal Constantin Mocanu, Raymond Veldhuis et al.2022 · 7 citationsarXiv
More ConvNets in the 2020s: Scaling up Kernels Beyond 51x51 using Sparsity2207.03620v3 · Shiwei Liu, Tianlong Chen, Xiaohan Chen et al.2022 · 85 citationsarXiv
Superposing Many Tickets into One: A Performance Booster for Sparse Neural Network Training2205.15322v3 · Lu Yin, Vlado Menkovski, Meng Fang et al.2022 · 2 citationsarXiv
The Unreasonable Effectiveness of Random Pruning: Return of the Most Naive Baseline for Sparse Training2202.02643v1 · Shiwei Liu, Tianlong Chen, Xiaohan Chen et al.2022 · 43 citationsarXiv
Avoiding Forgetting and Allowing Forward Transfer in Continual Learning via Sparse Networks2110.05329v3 · Ghada Sokar, Decebal Constantin Mocanu, Mykola Pechenizkiy2021 · 7 citationsarXiv
Evolving Plasticity for Autonomous Learning under Changing Environmental Conditions.34467993 · Yaman, Anil, Iacca, Giovanni, Mocanu, Decebal Constantin et al.2021 · 16 citationsEvolutionary computation. 2021;29(3):391-414
Deep Ensembling with No Overhead for either Training or Testing: The All-Round Blessings of Dynamic Sparsity2106.14568v6 · Shiwei Liu, Tianlong Chen, Zahra Atashgahi et al.2021 · 13 citationsProceedings of the International Conference on Machine Learning (ICLR 2022)
Sparse Training via Boosting Pruning Plasticity with Neuroregeneration2106.10404v4 · Shiwei Liu, Tianlong Chen, Xiaohan Chen et al.2021 · 47 citationsConference on Neural Information Processing Systems (NeurIPS 2021)
Dynamic Sparse Training for Deep Reinforcement Learning2106.04217v3 · Ghada Sokar, Elena Mocanu, Decebal Constantin Mocanu et al.2021 · 26 citationsarXiv
Sparse Training Theory for Scalable and Efficient Agents2103.01636v1 · Decebal Constantin Mocanu, Elena Mocanu, Tiago Pinto et al.2021 · 10 citations20th International Conference on Autonomous Agents and Multiagent Systems (AAMAS 2021)
Do We Actually Need Dense Over-Parameterization? In-Time Over-Parameterization in Sparse Training2102.02887v3 · Shiwei Liu, Lu Yin, Decebal Constantin Mocanu et al.2021 · 49 citationsProceedings of the 38th International Conference on Machine Learning (2021)
Truly Sparse Neural Networks at Scale2102.01732v2 · Selima Curci, Decebal Constantin Mocanu, Mykola Pechenizkiyi2021 · 7 citationsarXiv
Self-Attention Meta-Learner for Continual Learning2101.12136v1 · Ghada Sokar, Decebal Constantin Mocanu, Mykola Pechenizkiy2021 · 5 citations20th International Conference on Autonomous Agents and Multiagent Systems (AAMAS 2021)
Selfish Sparse RNN Training2101.09048v3 · Shiwei Liu, Decebal Constantin Mocanu, Yulong Pei et al.2021 · 21 citationsProceedings of the 38th International Conference on Machine Learning (2021)
Learning Invariant Representation for Continual Learning2101.06162v1 · Ghada Sokar, Decebal Constantin Mocanu, Mykola Pechenizkiy2021 · 10 citationsarXiv
Quick and Robust Feature Selection: the Strength of Energy-efficient Sparse Training for Autoencoders2012.00560v2 · Zahra Atashgahi, Ghada Sokar, Tim van der Lee et al.2020 · 10 citationsarXiv
SpaceNet: Make Free Space For Continual Learning2007.07617v3 · Ghada Sokar, Decebal Constantin Mocanu, Mykola Pechenizkiy2020 · 64 citationsNeurocomputing, 439: 1-11, 2021
Topological Insights into Sparse Neural Networks2006.14085v2 · Shiwei Liu, Tim Van der Lee, Anil Yaman et al.2020 · 24 citationsarXiv
Novelty Producing Synaptic Plasticity2002.03620v1 · Anil Yaman, Giovanni Iacca, Decebal Constantin Mocanu et al.2020 · 3 citationsarXiv
Softmax-based Classification is k-means Clustering: Formal Proof, Consequences for Adversarial Attacks, and Improvement through Centroid Based Tailoring2001.01987v1 · Sibylle Hess, Wouter Duivesteijn, Decebal Mocanu2020 · 8 citationsarXiv
On improving deep learning generalization with adaptive sparse connectivity1906.11626v1 · Shiwei Liu, Decebal Constantin Mocanu, Mykola Pechenizkiy2019 · 7 citationsarXiv
A Brain-inspired Algorithm for Training Highly Sparse Neural Networks1903.07138v3 · Zahra Atashgahi, Joost Pieterse, Shiwei Liu et al.2019 · 27 citationsarXiv
Sparse evolutionary Deep Learning with over one million artificial neurons on commodity hardware1901.09181v4 · Shiwei Liu, Decebal Constantin Mocanu, Amarsagar Reddy Ramapuram Matavalam et al.2019 · 54 citationsarXiv
Intrinsically Sparse Long Short-Term Memory Networks1901.09208v1 · Shiwei Liu, Decebal Constantin Mocanu, Mykola Pechenizkiy2019 · 5 citationsarXiv
Author Correction: Decentralized dynamic understanding of hidden relations in complex networks.29712929 · Mocanu, Decebal Constantin, Exarchakos, Georgios, Liotta, Antonio2019 · 0 citationsScientific reports. 2018;8(1):7007
Limited Evaluation Cooperative Co-evolutionary Differential Evolution for Large-scale Neuroevolution1804.07234v2 · Anil Yaman, Decebal Constantin Mocanu, Giovanni Iacca et al.2018 · 19 citationsarXiv
Career total: 123 works. 57 are in this corpus.Showing the 50 most recent.
Profile built from the corpus for this byline.
Author records are still filling in while the Hub is in alpha. If this is your page, you'll be able to claim it soon. Spot a mistake? Tell us.