Zobrazeno 1 - 10
of 101
pro vyhledávání: '"van der Wilk, Mark"'
For many machine learning methods, creating a model requires setting a parameter that controls the model's capacity before training, e.g.~number of neurons in DNNs, or inducing points in GPs. Increasing capacity improves performance until all the inf
Externí odkaz:
http://arxiv.org/abs/2408.07588
Weight space symmetries in neural network architectures, such as permutation symmetries in MLPs, give rise to Bayesian neural network (BNN) posteriors with many equivalent modes. This multimodality poses a challenge for variational inference (VI) tec
Externí odkaz:
http://arxiv.org/abs/2408.05496
Autor:
Qing, Jixiang, Langdon, Becky D, Lee, Robert M, Shafei, Behrang, van der Wilk, Mark, Tsay, Calvin, Misener, Ruth
We consider the problem of optimizing initial conditions and timing in dynamical systems governed by unknown ordinary differential equations (ODEs), where evaluating different initial conditions is costly and there are constraints on observation time
Externí odkaz:
http://arxiv.org/abs/2406.02352
With the rise in engineered biomolecular devices, there is an increased need for tailor-made biological sequences. Often, many similar biological sequences need to be made for a specific application meaning numerous, sometimes prohibitively expensive
Externí odkaz:
http://arxiv.org/abs/2402.17704
Autor:
Ober, Sebastian W., Artemev, Artem, Wagenländer, Marcel, Grobins, Rudolfs, van der Wilk, Mark
Gaussian processes (GPs) are a mature and widely-used component of the ML toolbox. One of their desirable qualities is automatic hyperparameter selection, which allows for training without user intervention. However, in many realistic settings, appro
Externí odkaz:
http://arxiv.org/abs/2402.09849
Autor:
Folch, Jose Pablo, Tsay, Calvin, Lee, Robert M, Shafei, Behrang, Ormaniec, Weronika, Krause, Andreas, van der Wilk, Mark, Misener, Ruth, Mutný, Mojmír
Bayesian optimization is a methodology to optimize black-box functions. Traditionally, it focuses on the setting where you can arbitrarily query the search space. However, many real-life problems do not offer this flexibility; in particular, the sear
Externí odkaz:
http://arxiv.org/abs/2402.08406
We present a method for systematically evaluating the correctness and robustness of instruction-tuned large language models (LLMs) for code generation via a new benchmark, Turbulence. Turbulence consists of a large set of natural language $\textit{qu
Externí odkaz:
http://arxiv.org/abs/2312.14856
Autor:
Folch, Jose Pablo, Odgers, James, Zhang, Shiqiang, Lee, Robert M, Shafei, Behrang, Walz, David, Tsay, Calvin, van der Wilk, Mark, Misener, Ruth
Publikováno v:
NeurIPS 2023 Workshop on Adaptive Experimental Design and Active Learning in the Real World
There has been a surge in interest in data-driven experimental design with applications to chemical engineering and drug manufacturing. Bayesian optimization (BO) has proven to be adaptable to such cases, since we can model the reactions of interest
Externí odkaz:
http://arxiv.org/abs/2312.00622
We propose an approach to do learning in Gaussian factor graphs. We treat all relevant quantities (inputs, outputs, parameters, latents) as random variables in a graphical model, and view both training and prediction as inference problems with differ
Externí odkaz:
http://arxiv.org/abs/2311.14649
Convolutions encode equivariance symmetries into neural networks leading to better generalisation performance. However, symmetries provide fixed hard constraints on the functions a network can represent, need to be specified in advance, and can not b
Externí odkaz:
http://arxiv.org/abs/2310.06131