AI/ML Projects
1164 projects · Page 37 of 49
LLM-Based Code Generation: Correctness and Security Audit Study
Audit generated code for correctness and vulnerabilities across models and prompts. You will learn code evaluation methodology, vulnerability classes and mitigation patterns.…
Active Learning Pipelines for Cost-Efficient Image Labelling
Cut labelling cost by choosing which images humans should annotate next. You will learn query strategies, labelling economics and stopping criteria.…
Spiking Neural Networks for Event-Camera Gesture Recognition
Recognise gestures from event streams with spiking networks and compare energy use. You will learn neuromorphic methods, event data handling and efficiency measurement.…
Transformer Models for ECG Arrhythmia Classification
Classify arrhythmias with attention models and compare against strong CNN baselines. You will learn physiological time series, class imbalance and clinical metrics.…
Domain Adaptation for Crop Disease Models Across Regions
Study why crop models fail in new regions and adapt them with limited local labels. You will learn distribution shift, adaptation techniques and field validation.…
Hallucination in LLM Medical Advice: Evaluation and Guardrail Design
Measure hallucination in medical answers and test retrieval and refusal guardrails. You will learn safety evaluation, guardrail engineering and risk communication.…
Reinforcement Learning for Adaptive Traffic Signal Control
Train signal controllers in simulation and compare against actuated baselines. You will learn RL environment design, traffic KPIs and sim-to-real caveats.…
Energy-Efficient Inference: Quantisation and Pruning Trade-Off Study
Map the accuracy-energy frontier for vision and language models under compression. You will learn compression methods, energy measurement and deployment guidance.…
Multi-Agent LLM Systems for Software Testing and Bug Reproduction
Build agents that turn bug reports into reproducing tests, and measure success rates. You will learn agent evaluation, test generation and software engineering benchmarks.…
Neuro-Symbolic Reasoning for Mathematical Word Problems
Combine language models with symbolic execution to solve word problems verifiably. You will learn neuro-symbolic design, program synthesis and verified reasoning.…
Fairness Auditing of Hiring Algorithms with Bias Mitigation
Audit a screening model for group disparities and test pre/in/post-processing mitigations. You will learn fairness definitions, audit methodology and mitigation trade-offs.…
Diffusion-Based Image Restoration for Low-Light Photography
Restore low-light photos with diffusion priors and compare against classical enhancement. You will learn image restoration, perceptual metrics and failure analysis.…
Speech Emotion Recognition for Call-Centre Quality Analytics
Analyse call emotion trajectories and link them to resolution outcomes. You will learn paralinguistic features, conversation analytics and privacy handling.…
Protein Function Prediction with Protein Language Models
Predict protein functions from sequence embeddings and evaluate on held-out families. You will learn bioinformatics ML, embedding evaluation and leakage-aware splits.…
Satellite Image Super-Resolution for Land-Use Mapping
Upscale satellite imagery and test whether land-use classification actually improves. You will learn super-resolution evaluation, downstream task validation and geospatial practice.…
Domain Adaptation for Satellite Imagery — Framework and Benchmark Study
Study the problem, build the proposed method and evaluate it rigorously with drift Monitoring. You will learn error analysis, technical paper writing and reproducible experiment design.…
A Framework for Retrieval-Augmented Support Automation with Multilingual Support
Study the problem, build the proposed method and evaluate it rigorously with multilingual Support. You will learn experimental design, ablation studies and error analysis.…
Speech Enhancement in Noisy Factory Audio — Comparative Evaluation
Study the problem, build the proposed method and evaluate it rigorously with drift Monitoring. You will learn benchmark design, statistical significance testing and baseline comparison.…
Contrastive Learning for Product Search: Methods, Design and Evaluation
Study the problem, build the proposed method and evaluate it rigorously with anomaly Flags. You will learn research methodology, experimental design and ablation studies.…
Active Learning for Annotation Budgets — Design and Evaluation
Study the problem, build the proposed method and evaluate it rigorously with explainable Outputs. You will learn ablation studies, statistical significance testing and experimental design.…
Handwriting Recognition for Historical Kannada Documents — Design and Evaluation
Study the problem, build the proposed method and evaluate it rigorously with batch Scoring. You will learn benchmark design, literature review methods and statistical significance testing.…
Bias Auditing of Screening Models — Framework and Benchmark Study
Study the problem, build the proposed method and evaluate it rigorously with anomaly Flags. You will learn reproducible experiment design, experimental design and error analysis.…
Graph-Based Fraud Ring: Methods, Design and Evaluation
Study the problem, build the proposed method and evaluate it rigorously with dashboard Visualisation. You will learn benchmark design, ablation studies and statistical significance testing.…
Retrieval-Augmented Support Automation — Design and Evaluation
Research, implement and evaluate the approach with benchmarks against strong baselines. You will learn statistical significance testing, research methodology and reproducible experiment design.…