Dr. Innocent Nyalala

Conference2025

Prime Directives for Responsible AI for Africa: A Manifesto for Inclusive Technology

Data Science Africa (DSA) 2025 conference, hosted by Iaban University

We present a holistic framework for developing inclusive, ethical, and sustainable AI that addresses Africa's unique cultural, social, and economic contexts. We emphasize participatory design, data equity, and local empowerment to ensure AI benefits all African communities.

Journal2025

Rectifying the extremely weakened signals for cassava leaf disease detection

Computers and Electronics in Agriculture, Volume 232

Developed an artificial intelligence approach to accurately detect cassava leaf diseases by enhancing weak signal patterns in agricultural imaging.

Dr. Manoj Kumar

Journal2025

Analysis of a Mathematical Model for Malaria Using a Data-Driven Approach

Scientific Reports, a journal published by Nature Portfolio

Our study, accepted in Scientific Reports, introduces a novel temperature and altitude-dependent SIR-SI model for malaria transmission. We integrate physics-informed neural networks (PINNs) for parameter inference and use Dynamic Mode Decomposition (DMD) to develop a real-time transmission risk index, offering a data-driven yet interpretable framework for predictive modeling and intervention planning.

Journal2025

Mathematical analysis and optimal control of age-structured social interaction model with law enforcement

Mathematical Methods in the Applied Sciences, Volume 48, Issue 3

Investigated the dynamics of social systems influenced by age groups and law enforcement through mathematical modeling.

Conference2025

Small but Mighty: Unlocking Financial Sentiment Analysis with Compact Language Models

Workshop on Advances in Financial Artificial Intelligence, International Conference on Learning Representations, Singapore

Proposed the use of compact natural language models for efficient financial sentiment analysis in resource-constrained environments.

Conference2024

Optimal control in a social interaction model with law enforcement

Control Conference Africa, Mauritius

Modeled how law enforcement interventions affect social behavior and optimal response strategies.

Invited Talk2024

Analysis of a mathematical model for malaria using data-driven approach

14th American Institute of Mathematical Sciences Conference, New York University Abu Dhabi

Demonstrated how data enhances predictive models for disease spread and control measures.

Dr. Tushar Shinde

Conference2025

Adaptive Compression of Large Vision Models for Efficient Image Quality Assessment of AI-Generated Content

IEEE/CVF International Conference on Computer Vision (ICCV) 2025 workshop on Efficient Computing under Limited Resources (ECLR), Honolulu, Hawaii, USA 🇺🇸.

Introduces a novel adaptive compression framework that enables large vision models to be used efficiently for assessing the visual quality of AI-generated content, paving the way for scalable, real-time deployment on resource-constrained platforms.

Conference2025

TWINSHIELD: A Dual Phase System for Detecting and Extracting Backdoors in VLMs

32th IEEE International Conference on Image Processing (ICIP) 2025, USA

Presents a dual-phase system to identify and remove backdoors in vision-language models (VLMs).

Conference2025

High-Performance Lightweight Vision Models for Land Cover Classification with Coresets and Compression

42nd International Conference on Machine Learning (ICML) 2025 workshop on TerraBytes, Canada

Proposes a high-efficiency model using coresets and compression for scalable land cover classification. Selected for the travel grant (registration support) from ICML Terrabytes 2025.

Conference2025

Towards Optimal Layer Ordering for Efficient Model Compression via Pruning and Quantization

25th IEEE International Conference on Digital Signal Processing (DSP) 2025, Greece

Explores the impact of layer ordering on model compression using pruning and quantization strategies.

Conference2025

EdgeHeart: Real-Time Detection of Cardiac Conduction Disorders on Edge Devices

50th IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP) 2025 (Show and Tell Session), India

Proposes a real-time edge-deployable system for detecting cardiac conduction anomalies.

Conference2025

Uncertainty-guided Style-aware Probabilistic Perceptual Quality Assessment for AI-Generated Images

IEEE / CVF Computer Vision and Pattern Recognition Conference (CVPR) 2025 Workshop, USA

Introduces a probabilistic quality assessment model for generative images, guided by style and uncertainty. Shivaanee, selected for the WiCV@CVPR 2025 travel grant.

Conference2025

An Efficient and Scalable Framework for Lightweight Crop Disease Recognition in Low-Resource Settings

IEEE / CVF Computer Vision and Pattern Recognition Conference (CVPR) 2025 Workshop on Agriculture-Vision, USA

Develops a lightweight and scalable framework for crop disease classification in resource-constrained environments.

Conference2025

Efficient Land-Cover Image Classification via Mixed Bit-Precision Quantization

International Conference on Learning Representations (ICLR) 2025 Workshop on Machine Learning for Remote Sensing, Singapore

Uses mixed-bit quantization for efficient land-cover image classification in remote sensing. Ahmed, selected for the €1500 travel grant from ICLR ML4RS 2025.

Conference2025

Model Compression Meets Resolution Scaling for Efficient Remote Sensing Classification

IEEE/CVF Winter Conference on Applications of Computer Vision (WACV) 2025 Workshop on Computer Vision for Earth Observation (CV4EO), USA

Integrates model compression and image resolution scaling for improved remote sensing classification.

Conference2025

Mixed-Precision is All You Need for Efficient Document Image Classification

IEEE/CVF Winter Conference on Applications of Computer Vision (WACV) 2025 Workshop on Computer Vision Systems for Document Analysis and Recognition, USA

Demonstrates the effectiveness of mixed-precision quantization for document image classification.

Conference2025

Navigating Coreset Selection and Model Compression for Efficient Maritime Image Classification

IEEE/CVF Winter Conference on Applications of Computer Vision (WACV) 2025 Workshop on Maritime Computer Vision, USA

Combines coreset selection and compression techniques for maritime visual recognition tasks.

Conference2024

Adaptive Quantization and Pruning of Deep Neural Networks via Layer Importance Estimation

38th Annual Conference on Neural Information Processing Systems (NeurIPS) 2024 workshop on Machine Learning and Compression, Canada

Proposes adaptive pruning and quantization based on estimated importance of each DNN layer.

Conference2024

Lightweight Neural Networks for Speech Emotion Recognition using Layer-wise Adaptive Quantization

38th Annual Conference on Neural Information Processing Systems (NeurIPS) 2024 ENLSP workshop (PMLR 262:584–595), Canada

Introduces a lightweight SER model leveraging adaptive layer-wise quantization.

Conference2024

Adaptive Quantization of Deep Neural Networks via Layer Importance Estimation

International Conference on Computer Vision & Image Processing (CVIP) 2024, India

Presents a method for quantization based on importance estimation of DNN layers for efficient deployment.

Conference2024

Adaptive Pixel-based Efficient Motion Search for Surveillance Video Coding

IEEE International Conference on Image Processing Theory, Tools and Applications (IPTA) 2024, Morocco

Proposes an adaptive motion estimation technique for efficient surveillance video compression.

Dr. Ritika Jain

Conference2025

Effective diagnosis of sleep disorders with LightGBM-EOG-EEG

13th IEEE International Conference on Healthcare Informatics, Italy

Combined EEG and EOG modalities with machine learning for better sleep disorder classification.

Conference2025

Seizure detection from EEG signals using autoencoder-classifier model

25th International Conference on Digital Signal Processing, Greece

Designed a deep learning approach that uses signal transformation for early seizure detection in clinical settings.

Journal2024

Modality-specific feature selection for improved sleep staging

IEEE Journal of Biomedical and Health Informatics, Volume 28, Issue 2

Enhanced sleep stage classification by selecting relevant EEG and EOG features and improving temporal awareness.

Conference2024

Classification of insomnia using sleep microstructure patterns

46th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, Florida

Explored the role of cyclic alternating patterns in identifying insomnia-related disruptions.

Journal2024

Effective diagnosis of various sleep disorders using LEE classifier: LightGBM-EOG

IEEE Journal of Biomedical and Health Informatics, Volume 29, Issue 4

Introduced a novel LightGBM-based classifier utilizing EOG signals to achieve efficient and interpretable diagnostics for a range of sleep disorders.

Journal2024

Modality-specific feature selection, data augmentation, and temporal context for improved sleep staging

IEEE Journal of Biomedical and Health Informatics, Volume 28, Issue 2

Proposed a comprehensive framework that integrates feature selection across EEG and EOG modalities, temporal modeling, and data augmentation to enhance accuracy in automatic sleep stage classification.

Awards & Recognitions