Student Projects in LAS Group

We offer Semester Projects, Bachelor’s and Master’s theses in our group. Depending on your preference, there are opportunities for working on theory, methods, and applications. M.Sc. projects at LAS often result in publications at leading conferences.

---

A list of topics for which we are actively recruiting students is given below. If you don’t see a project below that fits well, but you are interested in the kind of research our lab does, feel free to still reach out. To learn more about the research done in the group, visit our recent publications. You can also learn more about research by individual group members.

---

If you are interested in working with us, you can send an application by clicking the button below. Make sure that your email includes: a résumé, a recent transcript of records, your intended start date, and we highly recommend that you mention the projects you are interested in, members of the group with whom you would like to work, or recent publications by the group relevant to your interests.

---

---

If you are a bachelors or masters student but not a student at ETH, please see the opportunity listed in Applications for Summer Research Fellowships.

---

---

Current Topics

The detailed project proposals can be downloaded only from the ETH domain.

Continual Safe Adaptation in Multi-agent Domains

How drivers from Zurich can safely adapt to driving culture in New York?

---

---

Keywords: Multi-agent systems, Reinforcement Learning, Constraints

---

Addressing Reward Hacking in RLHF with Causality

A deep study of the reward hacking problem from a causal perspective.

---

---

Keywords: AI Alignment, Reward Hacking, Causality

---

Improving the Ability of Large Language Models

We offer various topics aimed at improving LLMs’ reasoning abilities.

---

---

Keywords: Large Language Models, Active Learning, Meta Learning, Computational Efficiency

---

Learning World Models for Legged Locomotion

Learn robust, structured models for policy learning.

---

---

Keywords: Reinforcement Learning, Curriculum Learning, Active Learning, Open-ended Learning

---

Pushing the Limit of Quadruped Running Speed

with Autonomous Curriculum Learning

---

---

Keywords: Reinforcement Learning, Curriculum Learning, Active Learning, Open-ended Learning

---

Myopic Behavior in Goal-reaching Reinforcement Learning

Allow a goal-reaching policy to be as greedy as it can afford to be.

---

---

Keywords: Reinforcement Learning, Optimization

---

De Novo Molecular Design via Diffusion Bandit Optimization

Discover promising molecules via novel algorithms merging diffusion models and bandit optimization.

---

---

Keywords: Molecular Design, Bandit Optimization, Diffusion Models, Generative Models

---

Online Safe Locomotion Learning in the Wild

Run reinforcement learning on real robots

---

---

Keywords: Online Reinforcement Learning, Safety, Robot learning

---

Autonomous Curriculum Learning for Increasingly Challenging Tasks

Proposing problems at the same time as they are being solved.

---

---

Keywords: Curriculum Learning, Open-ended learning, Robot learning

---

Humanoid Locomotion Learning and Finetuning from Human Feedback

Learn and finetune robotic motions with sequence-conditioned reward models from human feedback.

---

---

Keywords: Reinforcement learning from human feedback, Self-supervised Learning, Robot learning

---

Safe guaranteed domain exploration with autonomous robots

Develop and deploy algorithms for safe exploration with non-linear dynamics and unknown objectives

---

---

Keywords: Gaussian Processes, Active learning, Bayesian optimization, Model predictive control

---

Non-Convex Reinforcement Learning via Submodular Optimization

Develop algorithms for decision-making beyond the limitations of classic reinforcement learning.

---

---

Keywords: Reinforcement learning, Non-Markovian Rewards, Statistics, Optimization

---

Online Fair Classification for Sequential Data

We aim to design online fair classifier for sequential data.

---

---

Keywords: Fairness, Online Classification

---

Generalization for Meta-Learning and Personalized Federated Learning

We aim to find the correct formulation for many important problems in meta-learning and design algorithms to solve the correct formulation.

---

---

Keywords: Generalization, Meta-Learning, Personalized Federated Learning

---

Bayesian Optimization with Privacy

Develop private algorithms for Bayesian Optimization and show that privacy does not come for free, it takes a toll on performance.

---

---

Keywords: Bayesian Optimization, Differential Privacy, Meta-Learning

---

Optimized Sampling and Reconstruction in NMR Spectroscopy

Optimize state-of-art NMR machine with data-driven methods.  

---

Keywords: active learning, experiment design, real-world applications, neural networks

---

Point Processes for Species Modelling with Active Learning Citizen Science

Modelling species habitat using Point process modelling and active learning via citizen science

---

Keywords: active learning, experiment design, real-world applications, neural networks

---

Automating Biology with ML: Guiding Generative Modelling for Improved Protein Design

---

Keywords: generative modelling, proteins, enzymes, active learning, experiment design, real-world applications

---

Human kernels – querying similarity

What are the right assumptions before using ML? Sometimes we don’t know we know. Can machines help us?

---

Keywords: active learning, experiment design, real-world applications, neural networks

---

Graph Neural Optimization for Molecular Design

Employing Graph Neural Networks, develop a scalable Bayesian optimization algorithm which generates valid molecules with desirable profiles. [At capacity for Spring semester 2024.]

---

---

Keywords: Molecular Design, Energy-based Generative Models, Bayesian Optimization, Graph Neural Networks, Methodology, Applied

---

Machine Learning for Population Dynamics

Design and model spatio-temporal population dynamics using recent techniques in optimal transport and machine learning with focus on applications in single-cell biology.

---

Keywords: optimal transport, spatio-temporal dynamics, partial and stochastic difference equations

---

Structured Exploration in Large-Scale Sequential Decision-Making

How can we leverage structure for efficient exploration? How can we scale these techniques in the context of deep learning and large data?

---

Keywords: exploration, information-directed sampling, reinforcement learning, active learning

---

Confident Estimation via Online Convex Optimization

How can we make confident predictions using online convex optimization?

---

Keywords: online convex optimization, confidence sets, frequentist and Bayesian statistics

---

Applications of Machine Learning for Choosing Crop Varieties

Learning crop variety selection and management policies from data.

---

Keywords: applied, uncertainty quantification, active learning, reinforcement learning, remote sensing

---

Assimilation of crop growth models with remote sensing

Monitoring staple crops with satellite data.

---

Keywords: applied, remote sensing, sustainable agriculture

---

Machine Learning for Converter Control

Algorithms for control of power electronics converters, in collaboration with Hitachi Energy.

---

Keywords: applied, reinforcement learning, control    

General Areas

We offer projects in several general areas.

---

---

Examples of Previous Master Theses

---

---

---

---