Share This Author
Deep Learning Through the Lens of Example Difficulty
A measure of the computational difficulty of making a prediction for a given input: the (effective) prediction depth is introduced and surprising yet simple relationships between the prediction depth of a giveninput and the model’s uncertainty, confidence, accuracy and speed of learning for that data point are revealed.
What Do Neural Networks Learn When Trained With Random Labels?
It is shown analytically for convolutional and fully connected networks that an alignment between the principal components of network parameters and data takes place when training with random labels, and how this alignment produces a positive transfer.
Determining pressure-temperature phase diagrams of materials
© 2016 American Physical Society. ©2016 American Physical Society. We extend the nested sampling algorithm to simulate materials under periodic boundary and constant pressure conditions, and show how…
Structural Topology Optimization of Braced Steel Frameworks Using Genetic Programming
The bracing design for a three-bay, 12-story steel framework provides a preliminary test problem, giving promising initial results that reduce the structural mass of the bracing in comparison to previous published benchmarks for a displacement constraint based on design codes.
Structural optimisation in building design practice - case-studies in topology optimisation of bracing systems
- R. Baldock
- 26 October 2007
EVOLVING OPTIMIZED BRACED STEEL FRAMEWORKS FOR TALL BUILDINGS USING MODIFIED PATTERN SEARCH
Direct search methods offer potential for rapid exploration of a design space, enabling novel and optimized designs to be generated. We present the use of modified pattern search for optimizing the…
Classical Statistical Mechanics with Nested Sampling
- R. Baldock
- Computer Science
- 18 November 2017
Constant-pressure nested sampling with atomistic dynamics.
- R. Baldock, N. Bernstein, K. Salerno, L. Pártay, Gábor Csányi
- PhysicsPhysical review. E
- 30 October 2017
This paper enhances the nested sampling algorithm by using all-particle moves: either Galilean Monte Carlo or the total enthalpy Hamiltonian Monte Carlo algorithm, introduced in this paper, and shows the utility of the algorithms by calculating the order-disorder phase transition of a binary Lennard-Jones model alloy.
Exploiting molecular dynamics in Nested Sampling simulations of small peptides
Bayesian Neural Networks at Finite Temperature
We recapitulate the Bayesian formulation of neural network based classifiers and show that, while sampling from the posterior does indeed lead to better generalisation than is obtained by standard…