… With an epinet, conventional neural networks outperform large ensembles of hundreds or …
framework of Bayesian neural networks, so we introduce the epistemic neural network (ENN…

… In this paper, we build on the recent development of epistemic neural networks (ENNs), which
can offer a new and more effective approach for neural networks that know what they don’t …

… biases softmax neural nets trained with maximum likelihood reliably capture epistemic
uncertainty … We evaluate the epistemic uncertainty quality on active learning and OoD detection, …

… Next, we provide a quick overview on epistemic neural networks, which can make joint
predictions without being Bayesian. Finally, we introduce the ENN-DQN variant that allows for an …

… Abstract—Despite the success of deep neural networks (DNN) … of aleatoric and epistemic
uncertainty while maintaining a … the modeling of epistemic uncertainty while providing an …

… to propose a new class of neural networks able to model epistemic learning in a random
set/… We call this approach epistemic deep learning, and argue that a deep neural network …

… Our study first challenges the idea that epistemic uncertainty … methods are effective for this
epistemic UQ with theoretical … neural network, as they are more representative of epistemic …

… neural network is presented using the rapid neural network paradigm. Unlike the conventional
neural network training that optimizes the weights and biases of all the network layers by …

… It is important to note that this makes the neural network not … Hence, we cannot trust neural
networks to represent the world … , final form, neural networks start to resemble natural rather …

… We formally define the different sources of aleatoric and epistemic uncertainty in GNNs. …
We quantify the epistemic uncertainty arising from probabilistic parameters of the GNN …