Sökning: "Metropolis Hastings"
Visar resultat 1 - 5 av 16 uppsatser innehållade orden Metropolis Hastings.
1. Simuleringsdriven inferens av stokastiska dynamiska system
Kandidat-uppsats, Göteborgs universitet/Institutionen för matematiska vetenskaperSammanfattning : Stokastiska modeller, som ger tillförlitlig och användbar information om ett systems beteende, består ofta av stokastiska differentialekvationer (SDE) vars likelihoodfunktion inte är analytiskt tillgänglig. Mer traditionella Markov Chain Monte Carlo-metoder (MCMC) samt relativt nyligen utvecklade likelihood-fria Approximate Bayesian Computation-metoder (ABC) utgör populära angrepssätt för att utföra inferens på dessa typer av problem. LÄS MER
2. Monte Carlo study of disorder in electronic tight-binding models
Kandidat-uppsats, Lunds universitet/Fysiska institutionen; Lunds universitet/Matematisk fysikSammanfattning : Oftentimes solids are described by uniform, periodic lattices. In reality, however, there is often some disorder on some of the sites in the lattice. This disorder may come from, for example, there being a different type of atom or tightly bound electrons resulting in a larger on-site potential, affecting the electronic properties of the lattice. LÄS MER
3. Optimization and Bayesian Modeling of Road Distance for Inventory of Potholes in Gävle Municipality
Kandidat-uppsats, Stockholms universitet/Statistiska institutionenSammanfattning : Time management and distance evaluation have long been a difficult task for workers and companies. This thesis studies 6712 pothole coordinates in Gävle municipality, and evaluates the minimal total road distance needed to visit each pothole once, and return to an initial pothole. LÄS MER
4. Towards Deep Learning Accelerated Sparse Bayesian Frequency Estimation
Master-uppsats, Lunds universitet/Matematisk statistikSammanfattning : The Discrete Fourier Transform is the simplest way to obtain the spectrum of a discrete complex signal. This thesis concerns the case when the signal is known to contain a small (unknown) number of frequencies, not limited to the discrete Fourier frequencies, embedded in complex Gaussian noise. LÄS MER
5. Error estimation for neural network approximations of convection problems
Master-uppsats, KTH/Matematik (Avd.)Sammanfattning : The problem of approximating a solution to the convection equation $\partial_t u +f\cdot\nabla_xu = h$ given data on the flux $f$, source function $h$ and a final condition $g$ is investigated. Specifically, two layer neural networks are used to approximate $f,h$ and $g$ and a solution is approximated using numerical integration. LÄS MER