What is he best known for?
The fields of study he is best known for:
- Artificial intelligence
- Statistics
- Machine learning
His scientific interests lie mostly in Artificial intelligence, Pattern recognition, Curvature, Generative model and Inference.
Many of his studies involve connections with topics such as Machine learning and Artificial intelligence.
His Pattern recognition research is multidisciplinary, relying on both Autoencoder, Extrapolation and Nonparametric regression.
Roger Grosse has researched Autoencoder in several fields, including Latent variable, Posterior probability, Density estimation and Total correlation.
His Generative model research integrates issues from Deep learning, Convolutional Deep Belief Networks, Deep belief network, Probabilistic logic and Unsupervised learning.
In Reinforcement learning, he works on issues like Mathematical optimization, which are connected to Sample, Stochastic gradient descent, Fisher information and Applied mathematics.
His most cited work include:
- Convolutional deep belief networks for scalable unsupervised learning of hierarchical representations (2006 citations)
- Isolating Sources of Disentanglement in Variational Autoencoders (412 citations)
- Importance Weighted Autoencoders (340 citations)
What are the main themes of his work throughout his whole career to date?
The scientist’s investigation covers issues in Artificial neural network, Artificial intelligence, Algorithm, Mathematical optimization and Applied mathematics.
His Stochastic gradient descent study in the realm of Artificial neural network connects with subjects such as Generalization.
His Artificial intelligence study incorporates themes from Machine learning and Pattern recognition.
His studies in Pattern recognition integrate themes in fields like Residual and Generative model.
His Algorithm study combines topics from a wide range of disciplines, such as Kronecker delta and Markov chain Monte Carlo.
His study focuses on the intersection of Probabilistic logic and fields such as Unsupervised learning with connections in the field of Deep belief network and Convolutional Deep Belief Networks.
He most often published in these fields:
- Artificial neural network (41.90%)
- Artificial intelligence (36.19%)
- Algorithm (25.71%)
What were the highlights of his more recent work (between 2019-2021)?
- Artificial neural network (41.90%)
- Artificial intelligence (36.19%)
- Generalization (9.52%)
In recent papers he was focusing on the following fields of study:
His primary scientific interests are in Artificial neural network, Artificial intelligence, Generalization, Algorithm and Computation.
His Artificial neural network research is multidisciplinary, incorporating perspectives in Stability, Regularization, Mathematical optimization and Dropout.
As part of one scientific family, Roger Grosse deals mainly with the area of Mathematical optimization, narrowing it down to issues related to the Hyperparameter, and often Jacobian matrix and determinant.
In general Artificial intelligence, his work in Active learning and Deep learning is often linked to Bayesian linear regression and Gaussian process linking many areas of study.
His Algorithm study integrates concerns from other disciplines, such as Information theory, Lossy compression and Generative grammar.
The various areas that he examines in his Computation study include MNIST database, Theoretical computer science and Transformer.
Between 2019 and 2021, his most popular works were:
- Picking Winning Tickets Before Training by Preserving Gradient Flow (80 citations)
- Understanding and mitigating exploding inverses in invertible neural networks (10 citations)
- INT: An Inequality Benchmark for Evaluating Generalization in Theorem Proving. (6 citations)
In his most recent research, the most cited papers focused on:
- Statistics
- Artificial intelligence
- Machine learning
Roger Grosse spends much of his time researching Generalization, Artificial neural network, Computation, Artificial intelligence and Multiplicative noise.
His study on Generalization is intertwined with other disciplines of science such as Inequality, Benchmark, Monte Carlo tree search, Algebra and Measure.
The concepts of his Artificial neural network study are interwoven with issues in Sample, Theoretical computer science, Invertible matrix and Inverse problem.
His work carried out in the field of Computation brings together such families of science as Stability, MNIST database and Generative grammar.
He is interested in Pruning, which is a field of Artificial intelligence.
Upper and lower bounds, Rate of convergence, Applied mathematics, Gradient method and Acceleration are fields of study that intersect with his Multiplicative noise research.
This overview was generated by a machine learning system which analysed the scientist’s
body of work. If you have any feedback, you can
contact us
here.
