Thesis Worker 30 hp - Adaptive Hierarchical Inference with Foundation models
Technology, Data & Digital · Data, AI & Analytics · Machine Learning · Data Science · Embedded Systems
In short
This is a 30 hp thesis position within TRATON GROUP, focusing on Adaptive Hierarchical Inference with Foundation models for commercial-vehicle edge-cloud systems. The project involves designing and evaluating an uncertainty-aware inference cascade across edge devices, vehicle HPUs, and cloud resources, with a strong emphasis on research and development.
Responsibilities
- Conduct a literature study on compact foundation models, hierarchical inference, uncertainty estimation and calibration, learning to defer, and adaptive resource-aware edge-cloud offloading.
- Select and deploy a compact open-source model on a representative edge or HPU platform, characterizing its quality, latency, throughput, memory footprint, model size, and energy consumption.
- Implement and evaluate a confidence or uncertainty-estimation method for the local model.
- Develop an adaptive threshold-learning or cloud-escalation strategy that jointly considers uncertainty, task quality, and other relevant factors like latency and operational cost.
- Compare local-only, remote-only, static-threshold, adaptive-routing, and retrospective-oracle baselines, reporting trials and uncertainty for various metrics.
- Deliver reproducible code, experiment configurations, documented datasets and models, analysis scripts, and a clear account of limitations and deployment assumptions.
Requirements
- Master's programmes in Data Science, Machine Learning, Computer Science, Embedded Systems, Electrical Engineering, Engineering Physics, Applied Mathematics, or a similar field.
- Strong Python skills and experience with a deep-learning framework are expected.
- Familiarity with foundation-model inference, uncertainty estimation or calibration, embedded or edge platforms, multimodal or time-series learning, or performance profiling is beneficial.
Desired Qualifications
- Optional extension: a theoretical performance model, regret analysis, or a clearly scoped multimodal experiment.
Benefits
- Thesis work is an excellent way to get closer to Scania and build relationships for the future.
- Many of today's employees began their Scania career with their degree project.
- Subject to result quality and confidentiality constraints, the work may also contribute to a scientific publication.
Skills
PythonDeep Learning
#Foundation Models#Hierarchical Inference#Edge Computing#Cloud Computing#Machine Learning#Data Science#Commercial Vehicles#IoT
About Scania Group
Scania is a world-leading provider of transport solutions committed to a better tomorrow. Our purpose is to drive the shift towards a sustainable transport system. In doing so, we are creating a world of mobility that’s better for business, society and our environment. Employing more than 50,000 people in about 100 countries, Scania’s research and development is concentrated in Sweden, while production takes place in Europe and South America.
Company
Scania GroupJob Posted
3 days ago
Employment Type
Temporary
Work mode
On Site
Experience Level
Entry
Locations
Södertälje, Sweden
Qualification
Master
Applicants
Be an early applicant