Master Thesis Defensive Active Learning for Autonomous Driving (f/m/d)
NO NEED TO PREDICT THE FUTURE. YOU CAN CREATE IT.
SHARE YOUR PASSION.
World-leading technologies don’t make it into a BMW until they’ve undergone one of the most challenging journeys imaginable. It takes dynamic teams with outstanding technical skills to take them from the drawing board to the road. That’s why our experts will treat you as part of the team from day one, encourage you to bring your own ideas to the table – and give you the opportunity to really show what you can do.
We, the BMW Group, offer you an interesting and varied master thesis in the area of active learning for autonomous driving with neural networks and its threat of adversarial attacks.
Modern computer vision algorithms mostly rely on convolutional neural networks (CNNs) as they have become the state-of-the-art over the past decade. The effectiveness of CNNs heavily depends on the data used train it before deployment. As labeled data grows over years it becomes harder to identify relevant corner cases. Active learning aims to leverage the models’ uncertainty, beside other techniques, to collect relevant data. Contrary, adversarial attacks aim to fool the CNN by changing the input, causing an imperceptible misclassification with high confidence of the CNN. Within the scope of this thesis, a defensive active learning method shall be developed to provide and collect relevant data but also provides security against malicious inputs.
What awaits you?
Research on state-of-the-art methods in literature on active learning and adversarial attacks.
Implementation experience by developing a new method for active learning with neural networks to efficiently detect scenes that increase the expressiveness of training data and increased resilience against malicious inputs.
Evaluating the success of the novel active learning method on publicly available research datasets, leveraging cutting-edge training infrastructure.
Engagement in a team with experience in publishing at international peer-reviewed conferences.
Presenting the results of the thesis using the scientific method, both in written and oral form.
Working in an international and diverse team of doctoral candidates and students at the Autonomous Driving Campus in Unterschleißheim.
Please note that you must ensure that the thesis is supervised by a university.
What should you bring along?
You are a master student approaching the end of your degree in computer science or related fields with focus on machine learning or artificial intelligence.
You have strong knowledge in computer vision concepts (e.g. semantic segmentation), convolutional neural networks and adversarial attacks.
You have very good programming skills in Python, TensorFlow and worked with modern programming environment tools such as Docker and Git.
You speak English fluently.
You are motivated at solving problems independently as well as sharing ideas and working in a team.
You are enthused by new technologies and an innovative environment? Apply now!
At the BMW Group, we see diversity and inclusion in all its dimensions as a strength for our teams. Equal opportunities are a particular concern for us, and the equal treatment of applicants and employees is a fundamental principle of our corporate policy. That is why our recruiting decisions are also based on personality, experience and skills.