A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.
A new method designs batches for Bayesian optimization more efficiently.
Batch Normalization (BN) uses mini-batch statistics to normalize the activations during training, introducing dependence between mini-batch elements. This dependency can hurt the performance if the mini-batch size is too small, or if the elements are correlated. Several alternatives, such as Batch Renormalization and G…
The study analyzes batched methods for early stopping in stochastic multi-armed bandits.
SNGM improves large-batch training accuracy.
Adaptive batch sizes improve local gradient methods in distributed training.
A new method SEBS optimizes SGD batch size for better performance.
AdAdaGrad optimizes batch sizes for deep learning models, reducing the generalization gap.
Adaptive SGD learns optimal batch size for strong convex functions.
When using active learning, smaller batch sizes are typically more efficient from a learning efficiency perspective. However, in practice due to speed and human annotator considerations, the use of larger batch sizes is necessary. While past work has shown that larger batch sizes decrease learning efficiency from a lea…
New analysis reveals batch size effects on stochastic conditional gradient methods.
New research shows many batch selection methods for training work just as well as full batch training.
In this paper, we develop a new accelerated stochastic gradient method for efficiently solving the convex regularized empirical risk minimization problem in mini-batch settings. The use of mini-batches is becoming a golden standard in the machine learning community, because mini-batch settings stabilize the gradient es…
iMondrian forest combines isolation forest and Mondrian forest for better anomaly detection.
In many applications of black-box optimization, one can evaluate multiple points simultaneously, e.g. when evaluating the performances of several different neural network architectures in a parallel computing environment. In this paper, we develop a novel batch Bayesian optimization algorithm --- the parallel knowledge…
Matching pursuit (MP) methods are a promising class of feature construction algorithms for value function approximation. Yet existing MP methods require creating a pool of potential features, mandating expert knowledge or enumeration of a large feature pool, both of which hinder scalability. This paper introduces batch…
Large-scale distributed training of deep neural networks suffer from the generalization gap caused by the increase in the effective mini-batch size. Previous approaches try to solve this problem by varying the learning rate and batch size over epochs and layers, or some ad hoc modification of the batch normalization. W…
Mini-batch stochastic gradient descent and variants thereof have become standard for large-scale empirical risk minimization like the training of neural networks. These methods are usually used with a constant batch size chosen by simple empirical inspection. The batch size significantly influences the behavior of the …
Training deep neural networks with Stochastic Gradient Descent, or its variants, requires careful choice of both learning rate and batch size. While smaller batch sizes generally converge in fewer training epochs, larger batch sizes offer more parallelism and hence better computational efficiency. We have developed a n…
Combines multi-fidelity and asynchronous batch methods for faster experimental design.
Mini-batch stochastic gradient descent (SGD) and variants thereof approximate the objective function's gradient with a small number of training examples, aka the batch size. Small batch sizes require little computation for each model update but can yield high-variance gradient estimates, which poses some challenges for…
We analyze the learning properties of the stochastic gradient method when multiple passes over the data and mini-batches are allowed. We study how regularization properties are controlled by the step-size, the number of passes and the mini-batch size. In particular, we consider the square loss and show that for a unive…
Proposes qPO, a new acquisition strategy for batched Bayesian optimization that maximizes the probability of including the optimum.
A simple method improves batch active learning without high compute.
Batch normalisation doesn't affect variational inference but fails for larger batch sizes.
The paper studies batch decompositions of random datasets with probabilistic similarity constraints.
New method for robust learning from batches, even adversarial ones.
Proposes m-POT to improve m-OT's misspecified mappings issue.
Classical stochastic gradient methods for optimization rely on noisy gradient approximations that become progressively less accurate as iterates approach a solution. The large noise and small signal in the resulting gradients makes it difficult to use them for adaptive stepsize selection and automatic stopping. We prop…
Large-batch SGD is important for scaling training of deep neural networks. However, without fine-tuning hyperparameter schedules, the generalization of the model may be hampered. We propose to use batch augmentation: replicating instances of samples within the same batch with different data augmentations. Batch augment…
Extends batch active learning to non-differentiable models.
New method improves model accuracy in Byzantine-robust distributed learning by optimizing batch size.
In this paper we study a family of variance reduction methods with randomized batch size---at each step, the algorithm first randomly chooses the batch size and then selects a batch of samples to conduct a variance-reduced stochastic update. We give the linear convergence rate for this framework for composite functions…
In this paper, we study a simple algorithm to construct asymptotically valid confidence regions for model parameters using the batch means method. The main idea is to cancel out the covariance matrix which is hard/costly to estimate. In the process of developing the algorithm, we establish process-level functional cent…
The most straightforward method to accelerate Stochastic Gradient Descent (SGD) computation is to distribute the randomly selected batch of inputs over multiple processors. To keep the distributed processors fully utilized requires commensurately growing the batch size. However, large batch training often leads to poor…
This paper unifies three regularization methods in batch reinforcement learning.
New Convolutional Unit improves Batch Whitening performance.
A new method normalizes activations to match batch normalization without batch dependence.
Optimizes query routing to LLMs under cost and resource constraints.
We study a new aggregation operator for gradients coming from a mini-batch for stochastic gradient (SG) methods that allows a significant speed-up in the case of sparse optimization problems. We call this method AdaBatch and it only requires a few lines of code change compared to regular mini-batch SGD algorithms. We p…
A batched Gaussian Process bandit optimization method achieves near-optimal regret bounds.
A new method for batch prediction sets in classification problems.
New method for identifying best arm in batched multi-armed bandit problems.
Stochastic momentum methods trade compute efficiency for serial runtime.
Data collection and labeling is one of the main challenges in employing machine learning algorithms in a variety of real-world applications with limited data. While active learning methods attempt to tackle this issue by labeling only the data samples that give high information, they generally suffer from large computa…
Zero-shot anomaly detection method using batch normalization.
The standard L-BFGS method relies on gradient approximations that are not dominated by noise, so that search directions are descent directions, the line search is reliable, and quasi-Newton updating yields useful quadratic models of the objective function. All of this appears to call for a full batch approach, but sinc…
Bayesian optimization method for permutations accelerates combinatorial search.
A new method for graph neural networks speeds up inference and training.