Moreover, start with making a framework on what meta data you need. For example, data such as company information, dates of collecting data, etc. In case you find later in your research that you lack some data, this framework makes it relatively easy to repair your mistakes. Fourthly, developing machine learning algorithms and being able to explain what actually happens in these algorithms requires some statistical and mathematical knowledge.
Make sure you read into what is happening in the algorithm, or make sure you have access to the right people to explain this to you. There are quite a few standardized packages for e.
Master's Thesis - Asphalt modelling by deep learning - BIT Technology Solutions
However, you may still be asked to explain what happens in the neural network. Lastly, when you compare your algorithm to other algorithms, it is important that you measure the performance of your model in a similar way as done in the measurement of the other models. Otherwise, your comparison may be inaccurate.
For example, F1 is generally considered to better measure the ability of the model to discriminate than e. Reading into the meaning of these measurements and choosing the appropriate one may be fundamental to your research.
Advisor: Sushma Grellscheid , Carl Jones. In this project, we want to investigate if deep learning methods can contribute in combinatorial optimization problems in particular if it can find any pattern among good solutions for a pickup and delivery problem and if so, then take advantage of it to lead the search toward even better solutions. Develop a Machine Learning based Hyper-heuristic algorithm to solve a pickup and delivery problem. 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. Post-quantum cryptography is a hot area in the recent cryptographic research, aiming for building secure public-key cryptosystems even if quantum computers are assumed. The basic problem is to study the distribution of the huge amount of data generated during the cryptanalysis process that are difficult to represent in an analytic expression. Springer, Cham, This makes them on the one hand potentially highly valuable for business applications, but on the other, if their reasoning cannot be understood by decision makers or outcomes cannot be explained to end-users, difficult to put to use.
PricewaterhouseCoopers wish to provide an MSc student with a dataset representing a real business case, where the aim is either classification or regression. The student should develop both advanced and explainable machine learning regression or classification models. The advanced model s can be e.
If you want to suggest your own topic, please contact Pekka Parviainen. Skip to main content. Learning and inference with large Bayesian networks Most learning and inference tasks with Bayesian networks are NP-hard. Task: Evaluate quality of large-scale learning or inference algorithms empirically.
Advisor: Pekka Parviainen. Sum-product networks Traditionally, probabilistic graphical models use a graph structure to represent dependencies and independencies between random variables.
- Weitere interessante Jobs.
- format of a research paper thesis!
- Master Thesis at ISMLL.
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. 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. Bayesian deep learning Standard deep neural networks do not quantify uncertainty in predictions.
Deep learning for combinatorial problems Deep learning is usually applied in regression or classification problems. Automatic hyperparameter selection for isomap Isomap is a non-linear dimensionality reduction method with two free hyperparameters number of nearest neighbors and neighborhood radius.
Advisor: Nello Blaser. Directed cycle finding Finding cycles in directed graphs is one of the subroutines in many algorithms for learning the structure of Bayesian networks. Notions of stability in machine learning In topological data analysis, the term stability usually means that the output of an algorithm changes little, when the input is perturbed.
Validate persistent homology Persistent homology is a generalization of hierarchical clustering to find more structure than just the clusters. Persistent homology benchmarks Persistent homology is becoming a standard method for analyzing data. Divisive covers Divisive covers are a divisive technique for generating filtered simplicial complexes. Reinforcement learning for sparsification Reinforcement learning has recently become a way to heuristically solve optimization problems.
Topology of encodings State of the art for natural language processing and facial recognition use vector embedding algorithms such as word2vec or Siamese networks. Density estimation of single cell mass cytometry data with generative models Mass cytometry can measure up to forty different parameters for millions of blood cells. 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.
Donate to arXiv
Dimensionality reduction with missing data Many dimensionality reduction methods assume that complete data is available. Simulated underwater environment and deep learning Using data from the Mareano surveys or the LoVe underwater observatory, create a simulator for underwater benthic i. Using the simulated data, train deep learning neural networks to: a recognize presence of specific objects b locate specific objects c segment specific objects Test the systems on real data and evaluate the results. Advisor: Ketil Malde.
Simulated 3D environment and fish species identification Using data from the Deep Vision trawl camera, implement a simulator that mimics the stereoscopic camera setup. Evaluating the effects and interaction of hyperparameters in convolutional neural networks Neural networks have many hyperparameters, including choice of activation functions, regularization and normalization, gradient descent method, early stopping, cost function, and so on. Machine teaching Advisor: Jan Arne Telle. Machine learning approaches toward personalized treatment of leukemia With new data on multiple omics level reveal more information on leukemia and the effect of drugs, there are new opportunities to tailor treatment to each individual patient.
Advisor: Inge Jonassen. Developing Bayesian network inference algorithms for modelling genetic effects on gene expression Advisor: Tom Michoel. Deep learning approaches in imaging genetics Advisor: Tom Michoel. 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.
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. Analysing Nanopore data Oxford Nanopore technologies recently revolutionised Next Generation Sequencing by developing a portable Long-read sequencing device. Advisor: Maximilian Krause. 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.
Deep learning for a pickup and delivery problem In this project, we want to investigate if deep learning methods can contribute in combinatorial optimization problems in particular if it can find any pattern among good solutions for a pickup and delivery problem and if so, then take advantage of it to lead the search toward even better solutions. Advisor: Ahmad Hemmati. Machine Learning based Hyperheuristic algorithm Develop a Machine Learning based Hyper-heuristic algorithm to solve a pickup and delivery problem.
Machine learning for solving satisfiability problems and applications in cryptanalysis Advisor: Igor Semaev.
Related master thesis machine learning
Copyright 2019 - All Right Reserved