Postgraduate Diploma in Hierarchical Learning and Transferable Policies
Online 800-hour Postgraduate Diploma in Hierarchical Learning and Transferable Policies. 8 modules covering ISO/PAS 8800:2024, ISO 21448:2022 (SOTIF) and PyTorch. Assigned tutor and open enrolment.
Take the controls.
Run a real engineering trade-off from this field. Move the sliders, watch the model react in real time and beat the challenge.
What you will be able to do
Understand the hierarchical structure of the driving task.
Apply hierarchical reinforcement learning algorithms.
Learn manoeuvre primitives from demonstrations.
Transfer policies across tasks, domains and vehicles.
Use meta-learning for rapid adaptation.
Employ world models and learned representations.
Ensure the safety of transferred hierarchical policies.
Who it's for
- Reinforcement learning researcher seeking to specialise in Hierarchical Learning and Transferable Policies.
- Machine learning engineer seeking to specialise in Hierarchical Learning and Transferable Policies.
- Decision and planning engineer seeking to specialise in Hierarchical Learning and Transferable Policies.
- Robotics engineer seeking to specialise in Hierarchical Learning and Transferable Policies.
- Data scientist seeking to specialise in Hierarchical Learning and Transferable Policies.
- Simulation engineer seeking to specialise in Hierarchical Learning and Transferable Policies.
- Graduates in engineering or sciences seeking to enter the automotive sector.
What you'll work with
- ISO/PAS 8800:2024
- ISO 21448:2022 (SOTIF)
- ISO 26262:2018
- Reglamento (UE) 2024/1689 (Ley de IA)
- ISO/IEC 23053:2022
- UL 4600
- ISO 34503:2023
8 modules · 800 h
1.1Tasks and subtasks1.2Strategic, tactical and operational levels1.3Temporal abstraction of actions1.4Options and semi-MDP1.5Reusable skills1.6Exploration and credit assignment1.7Examples in robotics and driving1.8Challenges of learned hierarchy1.9Designed versus learned hierarchies1.10Case study: decomposition of an urban route
2.1Options framework and option-critic2.2Feudal networks FuN2.3HIRO and off-policy correction2.4HAC and hindsight goals2.5Hindsight Experience Replay2.6Unsupervised skill discovery2.7DIAYN and diversity2.8Comparison of approaches2.9Option termination and its learning2.10Lab: option-critic in a lane-following environment
3.1Segmentation of demonstrations3.2Manoeuvre primitives3.3Goal-conditioned policies3.4High-level planner with skills3.5Skill latents3.6Composition of primitives3.7Segmented driving data3.8Evaluation of primitives3.9Dynamic movement primitives DMP3.10Practice: manoeuvre primitive library
4.1Transfer across tasks and domains4.2Fine-tuning and layer freezing4.3Shared representations4.4Policy distillation4.5Sim-to-real transfer4.6Transfer across vehicles4.7Negative transfer4.8Transfer metrics4.9Lightweight adapters and LoRA4.10Lab: fine-tuning across scenarios
5.1Multitask learning5.2MAML and rapid adaptation5.3Context-based meta-RL5.4Driving styles5.5New cities5.6Continual learning without forgetting5.7Foundation models for driving5.8Evaluation of adaptation5.9Gradient conflicts between tasks5.10Practice: rapid adaptation with MAML
6.1State representations6.2Latent world models6.3Dreamer and planning in imagination6.4BEV as an interface between levels6.5Label-free self-supervised learning6.6Vision-language models for decision-making6.7Interpretability of skills6.8Computational cost6.9Generative video world models6.10Lab: world model in a simulator
7.1Guarantees by hierarchical level7.2Verification of individual skills7.3Monitors at the tactical level7.4Validation after transfer7.5ODD in new domains7.6ISO/PAS 8800 requirements7.7Model lifecycle7.8Documentation and traceability7.9Detection of unsupported domain7.10Workshop: safety argument for a transferred skill
8.1Tasks and skills8.2Environment and data8.3Low-level control8.4High-level decision-making8.5Transfer strategy8.6Evaluation in a new domain8.7Safety analysis8.8Discussion of results8.9Report and presentation8.10Final review with the tutor
How you'll learn
100% online training on the SEIUM campus, with downloadable material, recorded classes and assigned technical tutor. No fixed timetables: students progress at their own pace within the intake, which allows it to be combined with work.
Each module combines theory, guided reading of the original standard and applied exercises. The working software is PyTorch, CARLA, MetaDrive, Stable-Baselines3, Weights & Biases, in free, academic or evaluation versions. Trials are studied using real procedures for learning manoeuvre primitives from demonstrations, transferring policies between MetaDrive scenarios and rapid adaptation to new driving styles. Assessment is continuous per module plus the final integrating project.
Build it end to end
Your final project follows the real engineering workflow, step by step, reviewed by your tutor.
Understand the hierarchical structure of the driving task.
Apply hierarchical reinforcement learning algorithms.
Learn manoeuvre primitives from demonstrations.
Transfer policies across tasks, domains and vehicles.
Use meta-learning for rapid adaptation.
Where this takes you
Requirements
- Prior technical training or professional experience in the sector. A university degree is not required.
- Internet connection and a computer capable of running the programme's software.
- Sufficient Spanish-language proficiency to follow technical regulatory texts.
Ready to start?
Send your application today. Our admissions team replies within 72 hours with your study plan and funding options.
Frequently asked questions
The programme comprises 800 hours in total, spread over 6 to 8 months. That is an average of 10 to 15 hours per week, compatible with a full-time job.
The programme works with PyTorch, CARLA, MetaDrive, Stable-Baselines3, Weights & Biases. Free, academic or evaluation versions are used, and the tutor indicates the free alternative where one exists.
A SEIUM own diploma upon passing the continuous assessment and the integrating project. It is in-house training, not an official Ministry qualification.
Yes. Modules passed in a course or postgraduate diploma in the same area can be credited towards the higher-level programme. Check the credit transfer policy when you enrol.
Related programmes
Postgraduate Diploma in Secure Partitioning, ASIL and Inter-VM Communication
Online 800-hour Postgraduate Diploma in Secure Partitioning, ASIL and Inter-VM Communication. 8 modules covering ISO 26262:2018, ISO/SAE 21434:2021 and Ansys medini analyze. Assigned tutor and open…
Postgraduate Diploma in C-V2X/ITS-G5 and Cooperative Perception at the Edge
Online 800 h Postgraduate Diploma in C-V2X/ITS-G5 and Cooperative Perception at the Edge. 8 modules covering ETSI EN 302 637-2 (CAM), ETSI TS 103 324 (CPM)…
Postgraduate Diploma in Dense LiDAR and 3D SLAM for Automotive
Online 800-hour Postgraduate Diploma in Dense LiDAR and 3D SLAM for Automotive. 8 modules covering IEC 60825-1:2014, ISO 26262:2018 and Open3D. Assigned tutor and open enrolment.
More in Autonomous, ADAS & Software-Defined Vehicle
All Autonomous, ADAS & Software-Defined Vehicle programmes → · How to apply · Fees & financing · Scholarships