What is he best known for?
The fields of study he is best known for:
- Machine learning
- Statistics
- Artificial intelligence
Mathematical optimization, Artificial intelligence, Convex optimization, Algorithm and Stochastic optimization are his primary areas of study.
His research in Mathematical optimization intersects with topics in Stochastic gradient descent, Learnability and Regret.
His research integrates issues of Machine learning, Theoretical computer science and Euclidean space in his study of Artificial intelligence.
The study incorporates disciplines such as Function and Greedy algorithm in addition to Convex optimization.
His Algorithm research is multidisciplinary, incorporating perspectives in Artificial neural network, Norm and Time series.
Ohad Shamir combines subjects such as Distributed algorithm, Type and Applied mathematics with his study of Stochastic optimization.
His most cited work include:
- Optimal distributed online prediction using mini-batches (430 citations)
- Making Gradient Descent Optimal for Strongly Convex Stochastic Optimization (379 citations)
- Stochastic Gradient Descent for Non-smooth Optimization: Convergence Results and Optimal Averaging Schemes (341 citations)
What are the main themes of his work throughout his whole career to date?
His primary areas of study are Mathematical optimization, Artificial intelligence, Algorithm, Machine learning and Applied mathematics.
His Mathematical optimization research integrates issues from Sampling, Stochastic gradient descent, Regret and Convex optimization.
Ohad Shamir has included themes like Theoretical computer science and Pattern recognition in his Artificial intelligence study.
Artificial neural network, Dimension, Lipschitz continuity and Kernel is closely connected to Function in his research, which is encompassed under the umbrella topic of Algorithm.
In the field of Machine learning, his study on Semi-supervised learning overlaps with subjects such as Collaborative filtering.
His Applied mathematics study incorporates themes from Regularization, Gradient descent, Optimization problem, Upper and lower bounds and Stationary point.
He most often published in these fields:
- Mathematical optimization (26.11%)
- Artificial intelligence (25.12%)
- Algorithm (18.23%)
What were the highlights of his more recent work (between 2017-2021)?
- Artificial neural network (12.32%)
- Applied mathematics (14.29%)
- Function (13.79%)
In recent papers he was focusing on the following fields of study:
Ohad Shamir mainly focuses on Artificial neural network, Applied mathematics, Function, Upper and lower bounds and Stochastic gradient descent.
His Artificial neural network research is multidisciplinary, relying on both Exponential growth, Algorithm, Bounded function and Theoretical computer science.
His Applied mathematics research includes elements of Regularization, Non convex optimization, Gradient descent, Norm and Stationary point.
His biological study spans a wide range of topics, including Manifold, Mathematical optimization, Grassmannian and Convex optimization.
Minimax is the focus of his Mathematical optimization research.
Ohad Shamir interconnects Optimization problem and Combinatorics in the investigation of issues within Stochastic gradient descent.
Between 2017 and 2021, his most popular works were:
- Size-Independent Sample Complexity of Neural Networks. (176 citations)
- Spurious Local Minima are Common in Two-Layer ReLU Neural Networks. (92 citations)
- On the Power and Limitations of Random Features for Understanding Neural Networks (53 citations)
In his most recent research, the most cited papers focused on:
- Statistics
- Machine learning
- Artificial intelligence
Ohad Shamir mainly investigates Artificial neural network, Applied mathematics, Upper and lower bounds, Algorithm and Stochastic gradient descent.
His study looks at the intersection of Artificial neural network and topics like Exponential growth with Gradient descent and Polynomial.
His Upper and lower bounds research incorporates themes from Logarithm, Stochastic optimization, Minimax and Convex optimization.
His Convex optimization study combines topics from a wide range of disciplines, such as Regularization, Convex function, Numerical analysis and Mathematical optimization.
In general Algorithm study, his work on Residual neural network, Residual and Linear prediction often relates to the realm of Network architecture, thereby connecting several areas of interest.
His work deals with themes such as Optimization problem, Stationary point and Maxima and minima, which intersect with Stochastic gradient descent.
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.
