Master's thesis topics

Below is a list of available Master's thesis topics in the machine learning group

Available topics

Learning Bayesian networks with ancestral constraints

Bayesian networks are probabilistic models that are used to represent multivariate distributions. The core of a Bayesian network is its structure, a directed acyclic graph (DAG), that expresses conditional independencies between variables.

Typically, the structure is learned from data. The problem is NP-hard and thus exact algorithms do not scale up and one often resorts to heuristics that do not give any quality guarantees.

In a recent paper [1], we presented studied the complexity of learning Bayesian networks under ancestral constraints (for example, when a partial order between variables is known). However, the paper is fully theoretical and we do not know whether the proposed algorithms are useful in practice.

Task: Implement some of the algorithms and assess its practical performance experimentally.

Juha Harviainen & Pekka Parviainen: On Tractability of Learning Bayesian Networks with Ancestral Constraints. AISTATS 2025.

https://arxiv.org/pdf/2509.05002

Advisor: Pekka Parviainen

Learning and inference with large Bayesian networks

Most learning and inference tasks with Bayesian networks are NP-hard. Therefore, one often resorts to using different heuristics that do not give any quality guarantees.

Task: Evaluate quality of large-scale learning or inference algorithms empirically.

Advisor: Pekka Parviainen

Bayesian Bayesian networks

The naming of Bayesian networks is somewhat misleading because there is nothing Bayesian in them per se; A Bayesian network is just a representation of a joint probability distribution. One can, of course, use a Bayesian network while doing Bayesian inference. One can also learn Bayesian networks in a Bayesian way. That is, instead of finding an optimal network one computes the posterior distribution over networks.

Task: Develop algorithms for Bayesian learning of Bayesian networks (e.g., MCMC, variational inference, EM)

Advisor: Pekka Parviainen

Large-scale (probabilistic) matrix factorization

The idea behind matrix factorization is to represent a large data matrix as a product of two or more smaller matrices.They are often used in, for example, dimensionality reduction and recommendation systems. Probabilistic matrix factorization methods can be used to quantify uncertainty in recommendations. However, large-scale (probabilistic) matrix factorization is computationally challenging.

Potential thesis topics in this area: a) Develop scalable methods for large-scale matrix factorization (non-probabilistic or probabilistic), b) Develop probabilistic methods for implicit feedback (e.g., recommmendation engine when there are no rankings but only knowledge whether a customer has bought an item)

Advisor: Pekka Parviainen

Bayesian deep learning

Standard deep neural networks do not quantify uncertainty in predictions. On the other hand, Bayesian methods provide a principled way to handle uncertainty. Combining these approaches leads to Bayesian neural networks. The challenge is that Bayesian neural networks can be cumbersome to use and difficult to learn.

The task is to analyze Bayesian neural networks and different inference algorithms in some simple setting.

Advisor: Pekka Parviainen

Deep learning for combinatorial problems

Deep learning is usually applied in regression or classification problems. However, there has been some recent work on using deep learning to develop heuristics for combinatorial optimization problems; see, e.g., [1] and [2].

Task: Choose a combinatorial problem (or several related problems) and develop deep learning methods to solve them.

References: [1] Vinyals, Fortunato and Jaitly: Pointer networks. NIPS 2015. [2] Dai, Khalil, Zhang, Dilkina and Song: Learning Combinatorial Optimization Algorithms over Graphs. NIPS 2017.

Advisors: Pekka Parviainen, Ahmad Hemmati

Estimating the number of modes of an unknown function

Mode seeking considers estimating the number of local maxima of a function f. Sometimes one can find modes by, e.g., looking for points where the derivative of the function is zero. However, often the function is unknown and we have only access to some (possibly noisy) values of the function.

In topological data analysis, we can analyze topological structures using persistent homologies. For 1-dimensional signals, this can translate into looking at the birth/death persistence diagram, i.e. the birth and death of connected topological components as we expand the space around each point where we have observed our function. These observations turn out to be closely related to the modes (local maxima) of the function. A recent paper [1] proposed an efficient method for mode seeking.

In this project, the task is to extend the ideas from [1] to get a probabilistic estimate on the number of modes. To this end, one has to use probabilistic methods such as Gaussian processes.

[1] U. Bauer, A. Munk, H. Sieling, and M. Wardetzky. Persistence barcodes versus Kolmogorov signatures: Detecting modes of one-dimensional signals. Foundations of computational mathematics17:1 - 33, 2017.

Advisors: Pekka Parviainen, Nello Blaser

Causal Abstraction Learning

We naturally make sense of the world around us by working out causal relationships between objects and by representing in our minds these objects with different degrees of approximation and detail. Both processes are essential to our understanding of reality, and likely to be fundamental for developing artificial intelligence. The first process may be expressed using the formalism of structural causal models, while the second can be grounded in the theory of causal abstraction [1].

This project will consider the problem of learning an abstraction between two given structural causal models. The primary goal will be the development of efficient algorithms able to learn a meaningful abstraction between the given causal models.

[1] Rubenstein, Paul K., et al. "Causal consistency of structural equation models." arXiv preprint arXiv:1707.00819 (2017).

Advisor: Fabio Massimo Zennaro

Learning Causal Abstracted Models

Causal models allow us to describe real world systems by describing objects and how these objects influence each other. One property of this approach is that we can describe the same system in multiple levels of detail. This can be helpful when we want to reduce a causal model so it becomes more manageable to use. Causal abstractions give us the tools to describe how to go from a detailed causal model to a smaller causal model.

This project will aim to automate the process of creating an abstraction from one causal model to a smaller one. Mainly we aim to efficiently learn an abstraction that is consistent with some input model.

Advisor: Willem Schooltink, Fabio Massimo Zennaro

Causal Bandits

"Multi-armed bandit" is an informal name for slot machines, and the formal name of a large class of problems where an agent has to choose an action among a range of possibilities without knowing the ensuing rewards. Multi-armed bandit problems are one of the most essential reinforcement learning problems where an agent is directly faced with an exploitation-exploration trade-off.

This project will consider a class of multi-armed bandits where an agent, upon taking an action, interacts with a causal system [1]. The primary goal will be the development of learning strategies that takes advantage of the underlying causal system in order to learn optimal policies in a shortest amount of time.

[1] Lattimore, Finnian, Tor Lattimore, and Mark D. Reid. "Causal bandits: Learning good interventions via causal inference." Advances in neural information processing systems 29 (2016).

Advisor: Fabio Massimo Zennaro

Reinforcement Learning for Computer Security

The field of computer security presents a wide variety of challenging problems for artificial intelligence and autonomous agents. Guaranteeing the security of a system against attacks and penetrations by malicious hackers has always been a central concern of this field, and machine learning could now offer a substantial contribution. Security capture-the-flag simulations are particularly well-suited as a testbed for the application and development of reinforcement learning algorithms [1].

This project will consider the use of reinforcement learning for the preventive purpose of testing systems and discovering vulnerabilities before they can be exploited. The primary goal will be the modelling of capture-the-flag challenges of interest and the development of reinforcement learning algorithms that can solve them.

[1] Erdodi, Laszlo, and Fabio Massimo Zennaro. "The Agent Web Model--Modelling web hacking for reinforcement learning." arXiv preprint arXiv:2009.11274 (2020).

Advisor: Fabio Massimo Zennaro, Laszlo Tibor Erdodi

Multilevel Causal Discovery

Modelling causal relationships between variables of interest is a crucial step in understanding and controlling a system. A common approach is to represent such relations using graphs with directed arrows discriminating causes from effects.

While causal graphs are often built relying on expert knowledge, a more interesting challenge is to learn them from data. In particular, we want to consider the case where data might have been collected at multiple levels, for instance, with sensor with different resolutions. In this project we want to explore how these heterogeneous data can help the process of inferring causal structures.

[1] Anand, Tara V., et al. "Effect identification in cluster causal diagrams." Proceedings of the 37th AAAI Conference on Artificial Intelligence. Vol. 82. 2023.

Advisor: Fabio Massimo Zennaro, Pekka Parviainen

Manifolds of Causal Models

Modelling causal relationships is fundamental in order to understand real-world systems. A common formalism is offered by structural causal models (SCMs) which represent these relationships graphical. However, SCMs are complex mathematical objects entailing collections of different probability distributions.

In this project we want to explore a differential geometric perspective on structural causal models [1]. We will model an SCM and the probability distributions it generates in terms of manifold, and we will study how this modelling encodes causal properties of interest and how relevant quantities may be computed in this framework.

[1] Dominguez-Olmedo, Ricardo, et al. "On data manifolds entailed by structural causal models." International Conference on Machine Learning. PMLR, 2023.

Advisor: Fabio Massimo Zennaro, Nello Blaser

Abstraction for Epistemic Logic

Weighted Kripke models constitute a powerful formalism to express the evolving knowledge of an agent; it allows to express known facts and beliefs, and to recursively model the knowledge of an agent about another agent. Moreover, such relations of knowledge can be given a graphical expression using suitable diagrams on which to perform reasoning. Unfortunately, such graphs can quickly become very large and inefficient to process.

This project consider the reduction of epistemic logic graph using ideas from causal abstraction [1]. The primary goal will be the development of ML models that can learn to output small epistemic logic graph still satisfying logical and consistency constraints.

[1] Zennaro, Fabio Massimo, et al. "Jointly learning consistent causal abstractions over multiple interventional distributions." Conference on Causal Learning and Reasoning. PMLR, 2023

Advisor: Fabio Massimo Zennaro, Rustam Galimullin

Finalistic Models

The behavior of an agent may be explained both in causal terms (what has caused a certain behavior) or in finalistic terms (what aim justifies a certain behaviour). While causal reasoning is well explained by different mathematical formalism (e.g., structural causal models), finalistic reasoning is still object of research.

In this project we want to explore how a recently-proposed framework for finalistic reasoning [1] may be used to model intentions and counterfactuals in a causal bandit setting, or how it could be used to enhance inverse reinforcement learning.

[1] Compagno, Dario. "Final models: A finalistic interpretation of statistical correlation." arXiv preprint arXiv:2310.02272 (2023).

Advisor: Fabio Massimo Zennaro, Dario Compagno

Automatic hyperparameter selection for isomap

Isomap is a non-linear dimensionality reduction method with two free hyperparameters (number of nearest neighbors and neighborhood radius). Different hyperparameters result in dramatically different embeddings. Previous methods for selecting hyperparameters focused on choosing one optimal hyperparameter. In this project, you will explore the use of persistent homology to find parameter ranges that result in stable embeddings. The project has theoretic and computational aspects.

Advisor: Nello Blaser

Topological Ancombs quartet

This topic is based on the classical Ancombs quartet and families of point sets with identical 1D persistence (https://arxiv.org/abs/2202.00577). The goal is to generate more interesting datasets using the simulated annealing methods presented in (http://library.usc.edu.ph/ACM/CHI%202017/1proc/p1290.pdf). This project is mostly computational.

Advisor: Nello Blaser

Persistent homology vectorization with cycle location

There are many methods of vectorizing persistence diagrams, such as persistence landscapes, persistence images, PersLay and statistical summaries. Recently we have designed algorithms to in some cases efficiently detect the location of persistence cycles. In this project, you will vectorize not just the persistence diagram, but additional information such as the location of these cycles. This project is mostly computational with some theoretic aspects.

Advisor: Nello Blaser

Divisive covers

Divisive covers are a divisive technique for generating filtered simplicial complexes. They original used a naive way of dividing data into a cover. In this project, you will explore different methods of dividing space, based on principle component analysis, support vector machines and k-means clustering. In addition, you will explore methods of using divisive covers for classification. This project will be mostly computational.

Learning Acquisition Functions for Cost-aware Bayesian Optimization

This is a follow-up project of an earlier Master thesis that developed a novel method for learning Acquisition Functions in Bayesian Optimization through the use of Reinforcement Learning. The goal of this project is to further generalize this method (more general input, learned cost-functions) and apply it to hyperparameter optimization for neural networks.

Advisors: Nello Blaser, Audun Ljone Henriksen

Stable updates

This is a follow-up project of an earlier Master thesis that introduced and studied empirical stability in the context of tree-based models. The goal of this project is to develop stable update methods for deep learning models. You will design sevaral stable methods and empirically compare them (in terms of loss and stability) with a baseline and with one another.

Advisors: Morten Blørstad, Nello Blaser

Multimodality in Bayesian neural network ensembles

One method to assess uncertainty in neural network predictions is to use dropout or noise generators at prediction time and run every prediction many times. This leads to a distribution of predictions. Informatively summarizing such probability distributions is a non-trivial task and the commonly used means and standard deviations result in the loss of crucial information, especially in the case of multimodal distributions with distinct likely outcomes. In this project, you will analyze such multimodal distributions with mixture models and develop ways to exploit such multimodality to improve training. This project can have theoretical, computational and applied aspects.

Advisor: Nello Blaser

Optimizing Jet Reconstruction with Quantum-Based Clustering Techniques

QCD jets are collimated sprays of energy and particles frequently observed at collider experiments, signaling the occurrence of high-energy processes. These jets are pivotal for understanding quantum chromodynamics at high energies and for exploring physics beyond the Standard Model. The definition of a jet typically arises from an agreement between experimentalists and theorists, formalized in jet algorithms that help make sense of the large number of particles produced in collisions.

This project focuses on jet reconstruction using data-driven clustering techniques. Specifically, we aim to apply fast clustering algorithms, optimized through quantum methods, to identify the optimal distribution of jets on an event-by-event basis. This approach allows us to refine jet definitions and enhance the accuracy of jet reconstruction. Key objectives include:

Introduce a purely data-drive clustering process using standard techniques.
Optimizing the clustering process using quantum-inspired techniques.
Benchmark the performance of these algorithms against existing frameworks and compare the extracted jet populations.

By focusing on clustering methods and quantum optimization, this project aims to provide a novel perspective on jet reconstruction, improving the precision and reliability of high-energy physics analyses.

Advisors: Nello Blaser, Konrad Tywoniuk

Precise Navigation of a Lego Robot

Precise navigation based only on local measurements is a challenging problem. This task appears in many important applications, including subsurface well-steering in lateral drilling operations and robot-driving in the First Lego League. In both tasks, the measurements are local, unlike global systems such as GPS. Logging-while-drilling gamma-ray measurements only give information about rock properties just outside the well. Color sensors on a Lego robot can only identify the average color on the play-mat underneath the robot. The positioning thus should be achieved by a probabilistic estimation of location using a reference "map" of geology / play-mat.

In this project, the task is to (1) develop a practical machine-learning positioning and navigation method and (2) test it on a Lego robot. Over the course of the project, the prototype should be transformed from in-silico to real-life testing on a prototype (read: you can play with Lego). For the latter, you will need to account for the onboard computational constraints of the Lego robot.

Reading related to the low-data well-steering problem:

[1] S Alyaev, AH Elsheikh (2022) “Direct multi‐modal inversion of geophysical logs using deep learning”. Earth and Space Science 9 (9), e2021EA002186.

[2] RB Muhammad, Y Cheraghi, S Alyaev, A Srivastava, RB Bratvold (2025). “Geosteering Robot Powered by Multiple Probabilistic Interpretations and Artificial Intelligence: Benchmarking Against Human Experts” SPE Journal 30 (03), 995-1009.

Advisors: Nello Blaser, Sergey Alyaev (NORCE)

Classification using Multiparameter Persistence

Multiparameter persistence is a fast-growing field within topological data analysis. Typically, the extra parameters give a more robust construction that better handles noise compared to their one-parameter counterparts. In recent years, several two-parameter filtrations have been introduced, along with several topological descriptors and ways of vectorizing these.

This project will consider different combinations of filtrations and descriptors, and compare their efficiency on point cloud classification.

Useful references include the work by Scoccola et al. https://arxiv.org/pdf/2406.07224 giving a general framework for studying differentiation of a range of descriptors, and the papers of Carrière and Blumberg https://web.ma.utexas.edu/users/blumberg/camera-ready.pdf and Vipond https://jmlr.org/papers/volume21/19-054/19-054.pdf as examples of descriptors for multiparameter persistence used in machine learning.

Advisor: Lars Moberg Salbu

Multipersistence Based Clustering Using the Core Bifiltration

In topological data analysis, one-parameter filtrations like the Vietoris-Rips complex are overly sensitive to outliers. A more robust approach is to add a density parameter to the filtration, leading to multiparameter persistence.

Rolle and Scoccola https://www.jmlr.org/papers/volume25/21-1185/21-1185.pdf recently developed a stable clustering method using used the two-parameter filtration degree-Rips, showcasing one way multiparameter persistence can be applied to machine learning.

This project will combine the work of Rolle and Scoccola, with our recent work on the Core bifiltration https://arxiv.org/pdf/2405.01214, a two-parameter filtration that is both stable and fast to compute for lower dimensions. The goal is to obtain a robust density-based clustering method that is fast to compute.

Advisor: Lars Moberg Salbu

Learning Point Clouds From Persistence Diagrams

The persistence diagram is an important tool in topological data analysis. It describes the evolving topology of a space through a filtration. Typically we consider combinatorial spaces that are constructed from point cloud data, like the Vietoris-Rips complex or \v Cech complex, or more generally from numerical tabular data like the Dowker complex.

A natural question is how much information the persistence diagrams retains from the original data. This project concerns exactly this reverse problem of learning a point cloud (or tabular data) from their persistence diagram.

Relevant literature includes the work of Carrière et al. https://proceedings.mlr.press/v139/carriere21a/carriere21a.pdfstudying the differentiability of calculating the persistence diagram.

Advisor: Lars Moberg Salbu

Online learning in real-time systems

Build a model for the drilling process by using the Virtual simulator OpenLab (https://openlab.app/) for real-time data generation and online learning techniques. The student will also do a short survey of existing online learning techniques and learn how to cope with errors and delays in the data.

Advisor: Rodica Mihai

Building a finite state automaton for the drilling process by using queries and counterexamples

Datasets will be generated by using the Virtual simulator OpenLab (https://openlab.app/). The student will study the datasets and decide upon a good setting to extract a finite state automaton for the drilling process. The student will also do a short survey of existing techniques for extracting finite state automata from process data. We present a novel algorithm that uses exact learning and abstraction to extract a deterministic finite automaton describing the state dynamics of a given trained RNN. We do this using Angluin's L*algorithm as a learner and the trained RNN as an oracle. Our technique efficiently extracts accurate automata from trained RNNs, even when the state vectors are large and require fine differentiation.arxiv.org

Advisor: Rodica Mihai

Space-Time Linkage of Fish Distribution to Environmental Conditions

Background

Conditions in the marine environment, such as, temperature and currents, influence the spatial distribution and migration patterns of marine species. Hence, understanding the link between environmental factors and fish behavior is crucial in predicting, e.g., how fish populations may respond to climate change. Deriving this link is challenging because it requires analysis of two types of datasets (i) large environmental (currents, temperature) datasets that vary in space and time, and (ii) sparse and sporadic spatial observations of fish populations.

Project goal

The primary goal of the project is to develop a methodology that helps predict how spatial distribution of two fish stocks (capelin and mackerel) change in response to variability in the physical marine environment (ocean currents and temperature). The information can also be used to optimize data collection by minimizing time spent in spatial sampling of the populations.

Approach

The project will focus on the use of machine learning and/or causal inference algorithms. As a first step, we use synthetic (fish and environmental) data from analytic models that couple the two data sources. Because the ‘truth’ is known, we can judge the efficiency and error margins of the methodologies. We then apply the methodologies to real world (empirical) observations.

Advisors: Tom Michoel, Sam Subbey.

Towards precision medicine for cancer patient stratification

On average, a drug or a treatment is effective in only about half of patients who take it. This means patients need to try several until they find one that is effective at the cost of side effects associated with every treatment. The ultimate goal of precision medicine is to provide a treatment best suited for every individual. Sequencing technologies have now made genomics data available in abundance to be used towards this goal.

In this project we will specifically focus on cancer. Most cancer patients get a particular treatment based on the cancer type and the stage, though different individuals will react differently to a treatment. It is now well established that genetic mutations cause cancer growth and spreading and importantly, these mutations are different in individual patients. The aim of this project is use genomic data allow to better stratification of cancer patients, to predict the treatment most likely to work. Specifically, the project will use machine learning approach to integrate genomic data and build a classifier for stratification of cancer patients.

Advisor: Anagha Joshi

Unraveling gene regulation from single cell data

Multi-cellularity is achieved by precise control of gene expression during development and differentiation and aberrations of this process leads to disease. A key regulatory process in gene regulation is at the transcriptional level where epigenetic and transcriptional regulators control the spatial and temporal expression of the target genes in response to environmental, developmental, and physiological cues obtained from a signalling cascade. The rapid advances in sequencing technology has now made it feasible to study this process by understanding the genomewide patterns of diverse epigenetic and transcription factors as well as at a single cell level.

Single cell RNA sequencing is highly important, particularly in cancer as it allows exploration of heterogenous tumor sample, obstructing therapeutic targeting which leads to poor survival. Despite huge clinical relevance and potential, analysis of single cell RNA-seq data is challenging. In this project, we will develop strategies to infer gene regulatory networks using network inference approaches (both supervised and un-supervised). It will be primarily tested on the single cell datasets in the context of cancer.

Advisor: Anagha Joshi

Machine learning for preventive medicine

Advisor: Anagha Joshi

Developing a Stress Granule Classifier

To carry out the multitude of functions 'expected' from a human cell, the cell employs a strategy of division of labour, whereby sub-cellular organelles carry out distinct functions. Thus we traditionally understand organelles as distinct units defined both functionally and physically with a distinct shape and size range. More recently a new class of organelles have been discovered that are assembled and dissolved on demand and are composed of liquid droplets or 'granules'. Granules show many properties characteristic of liquids, such as flow and wetting, but they can also assume many shapes and indeed also fluctuate in shape. One such liquid organelle is a stress granule (SG).

Stress granules are pro-survival organelles that assemble in response to cellular stress and important in cancer and neurodegenerative diseases like Alzheimer's. They are liquid or gel-like and can assume varying sizes and shapes depending on their cellular composition.

In a given experiment we are able to image the entire cell over a time series of 1000 frames; from which we extract a rough estimation of the size and shape of each granule. Our current method is susceptible to noise and a granule may be falsely rejected if the boundary is drawn poorly in a small majority of frames. Ideally, we would also like to identify potentially interesting features, such as voids, in the accepted granules.

We are interested in applying a machine learning approach to develop a descriptor for a 'classic' granule and furthermore classify them into different functional groups based on disease status of the cell. This method would be applied across thousands of granules imaged from control and disease cells. We are a multi-disciplinary group consisting of biologists, computational scientists and physicists.

Advisors: Sushma Grellscheid, Carl Jones

Machine Learning based Hyperheuristic algorithm

Develop a Machine Learning based Hyper-heuristic algorithm to solve combinatorial optimization problems. A hyper-heuristic is a heuristics that choose heuristics automatically. Hyper-heuristic seeks to automate the process of selecting, combining, generating or adapting several simpler heuristics to efficiently solve computational search problems [Handbook of Metaheuristics]. There might be multiple heuristics for solving a problem. Heuristics have their own strength and weakness. In this project, we want to use machine-learning techniques to learn the strength and weakness of each heuristic while we are using them in an iterative search for finding high quality solutions and then use them intelligently for the rest of the search. Once a new information is gathered during the search the hyper-heuristic algorithm automatically adjusts the heuristics.

Advisor: Ahmad Hemmati

Image calibration in images for size estimation and density measures of marine species

The Mareano Project at the Institute of Marine Research has been tasked with studying vulnerable, yet environmentally important seabed communities in Norwegian waters since 2006. A large component of this involves collecting video of the seafloor and identifying, quantifying and localizing the marine life within. However, image annotation is a labour-intensive task that introduces inconsistencies and human error. The following projects aim to develop automated tools to improve the extraction and quality of marine ecological data from imagery.

Estimating the size of marine species is important for numerous reasons such as measuring ecosystem health, biodiversity and recovery to support conservation and management. In underwater image surveys, accurate calculation of image area is crucial to this and complicated by the variation in altitude and angle of camera platforms in relation to the seabed. Lasers, fixed at a set distance, are required for scaling. However, the pixel distance between lasers is measured manually by analysts and the visibility of the lasers varies.

This project will look to develop a pipeline that will first detect lasers in imagery and highlight if lasers are missing or appearing inconsistently. The image area can then be estimated, accounting for angle of the seabed. With a known image area, object detection/segmentation methods will be explored to automatically estimate the size/density of marine species. Other emerging methods for image calibration can be explored to improve practical applicability, speed and efficiency.

Advisors: Chloe Game (UIB) and Nils Piechaud (Institute of Marine Research)

Tracking of marine species in video for optimal frame extraction

Monitoring of the deep sea and species living on the seabed requires ecologists to manually extract representative and high-quality frames from video for their analysis. This can be a challenging process given the technical difficulty with imaging in these environments. Images suffer from effects such as blurring, low contrast, non-uniform illumination, colour saturation and loss and thus the appearance of species is highly variable. This is exacerbated further by the uneven movement of the imaging platform in 3D space.

This project will look to develop a pipeline to automatically detect and track seabed species over multiple video frames improving the manual video annotation procedure. This will be used to design a tool to prepare image datasets from video in a standardized, representative way that avoids duplication and seeks high quality imagery of marine species of interest that can be used for further ecological analysis and generating training datasets for machine learning tasks.

Advisors: Chloe Game (UIB) and Nils Piechaud (Institute of Marine Research)

Machine learning for solving satisfiability problems and applications in cryptanalysis

Advisor: Igor Semaev

Backpropagation in branching stochastic processes

Branching stochastic processes are mathematical models for self-replicating entities in biology (which could be cells, individuals in a population, or distinct species depending on the context). In such processes individuals have one or more internal degrees of freedom whose dynamics is described by a stochastic differential equation (SDE). After a certain lifetime, an individual dies and splits into multiple independent offspring individuals. This process is repeated indefinitely, producing a collection of N(t) individuals at time t. To fit such models to data, researchers have traditionally relied on mathematically tractable, but biologically simplistic assumptions about the underlying dynamics. However, the emergence of "scientific machine learning" (SciML), which combines principles from machine learning (composability, automatic differentiation, neural nets as universal function approximators, etc.) with classical dynamical systems modelling, is expanding the traditional notion of what it means for a model to be "tractable". A candidate idea for applying SciML to learn more realistic branching processes with non-linear dynamics would be to combine backpropagation through the lineage tree to infer the hidden states of extinct ancestors with forward simulation of neural SDEs. The aim of this project is to develop the theoretical basis for this and related ideas, and use the new models and algorithms to learn kinetic parameters and causal interactions from clonal gene expression data.