Research
On-device research index

arXiv research

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.

168,695 papers · 148 categories

Trend · papers per month

138276413551 · Jun 202019922001200920172026
48 results for large activation

Large deviation principle for deep neural networks with ReLU activation.

problem Understanding the behavior of deep neural networks with ReLU activation.
method Proving a large deviation principle for networks with Gaussian weights and ReLU activation functions.
result Simplified expressions and power-series expansions for the ReLU case.

This paper studies activation sparsity in large language models, finding key trends and implications.

problem Activation sparsity in large language models (LLMs) can be improved for efficiency and interpretability.
method Proposes PPL-p%p\% sparsity, analyzes trends with training data, width-depth ratio, and parameter scale.
result ReLU is more efficient for sparsity than SiLU, and deeper architectures can improve sparsity.

Active testing for large language models is made more efficient and accurate.

problem Efficient evaluation of large language models with limited labels.
method Cost-saving measures and in-context learning for constructing a surrogate model.
result Significantly more accurate evaluations of LLM performance compared to random data acquisition.

Paper proposes a new activation function to reduce overfitting and large weight update issues.

problem Overfitting and large weight update problems in neural networks.
method Introduces a new activation function called Thresholded Exponential Rectified Linear Units (TERELU).
result TERELU shows better performance in reducing overfitting and large weight update issues compared to other activation functions.

Deep learning models can overfit noisy data without losing generalization.

problem Understanding the generalization of deep learning models in noisy data.
method Empirical investigation of epoch-wise double descent in fully connected neural networks trained on CIFAR-10 with 30% label noise.
result The model achieves strong re-generalization on test data after overfitting noisy training data, corresponding to a 'benign overfitting' state.

Efficiently selects nearest neighbors for labeling to speed up active learning.

problem Intractable active learning and search for large-scale unlabeled data.
method Restricts candidate pool to nearest neighbors of labeled set.
result Achieved similar performance to global approach but reduced computational cost by up to 3 orders of magnitude.

S2D selectively decays large singular values to improve quantization of neural activations.

problem Large activation outliers in transformer models cause accuracy drops during quantization.
method Selective Spectral Decay (S2DS^2D) that surgically regularizes only the largest singular values.
result Significantly reduces activation outliers and produces well-conditioned representations.

We propose a new batch mode active learning algorithm designed for neural networks and large query batch sizes. The method, Discriminative Active Learning (DAL), poses active learning as a binary classification task, attempting to choose examples to label in such a way as to make the labeled set and the unlabeled pool …

2019-07-15abs ↗pdf ↗

Active testing reduces label costs for efficient model evaluation.

problem Real-world applications require expensive test labels, disconnecting from existing model evaluation methods.
method Derives acquisition strategies to select test points efficiently, addressing label bias and variance.
result Active testing improves model evaluation efficiency without sacrificing accuracy.

Annotating the right data for training deep neural networks is an important challenge. Active learning using uncertainty estimates from Bayesian Neural Networks (BNNs) could provide an effective solution to this. Despite being theoretically principled, BNNs require approximations to be applied to large-scale problems, …

2018-11-08abs ↗pdf ↗

Convolutional neural networks (CNNs) have been successfully applied to many recognition and learning tasks using a universal recipe; training a deep model on a very large dataset of supervised examples. However, this approach is rather restrictive in practice since collecting a large set of labeled images is very expen…

2017-08-01abs ↗pdf ↗

FAQ efficiently evaluates LLMs with statistical guarantees using adaptive query selection.

problem Efficiently evaluating many LLMs on a large suite of benchmarks is expensive.
method FAQ uses Bayesian factor models, adaptive sampling, and proactive active inference to select queries.
result FAQ delivers up to 5x effective sample size gains over baselines, matching CI width with fewer queries.

IMPACT optimizes LLM compression by focusing on activation importance, reducing model size up to 55.4%.

problem Resource constraints in deploying large language models (LLMs).
method IMPACT integrates activation importance into low-rank compression, optimizing for both size and accuracy.
result IMPACT achieves up to 55.4% greater model size reduction while maintaining comparable or better accuracy.

Bal-PM reduces preference labeling costs for LLMs.

problem Efficiently acquiring human feedback for preference modeling in large language models.
method Bayesian Active Learning with entropy maximization in feature space.
result Bal-PM reduces the number of required preference labels by 33% to 68%.

A new algorithm improves efficiency in selecting examples for deep learning.

problem Efficiently choosing multiple examples to mark up for deep learning on large datasets.
method Large BatchBALD algorithm, approximating BatchBALD with reduced computational complexity.
result Comparable quality in selection while significantly reducing computation time, especially for large batches.

Leveraging the wealth of unlabeled data produced in recent years provides great potential for improving supervised models. When the cost of acquiring labels is high, probabilistic active learning methods can be used to greedily select the most informative data points to be labeled. However, for many large-scale problem…

2019-08-06abs ↗pdf ↗

Improves decentralized learning by teleporting active nodes for better convergence.

problem Decentralized learning's convergence rate degrades with large node numbers.
method Activates a subset of nodes, fetches parameters from previous active nodes, updates, and performs gossip averaging on a small topology.
result Teleportation completely alleviates convergence rate degradation with proper node activation.

Neural networks learn distance-based representations, not just intensity.

problem Understanding how neural networks interpret and learn from internal activations.
method Manipulated ReLU and Absolute Value activations to observe sensitivity to distance and intensity perturbations.
result Neural networks are highly sensitive to small distance-based perturbations, challenging the intensity-based interpretation.

In the world of big data, large but costly to label datasets dominate many fields. Active learning, a semi-supervised alternative to the standard PAC-learning model, was introduced to explore whether adaptive labeling could learn concepts with exponentially fewer labeled samples. While previous results show that active…

2019-07-08abs ↗pdf ↗

Improved EXACT strategy reduces GNN memory consumption and runtime.

problem Efficiently training large-scale GNNs with reduced memory usage.
method Block-wise quantization of intermediate activation maps with improved variance minimization.
result Further reduction in memory consumption (>15%) and runtime speedup (5%) with similar performance trade-offs.

Wide networks with polynomial activations have proven asymptotic behavior.

problem Understanding the behavior of neural networks in the large width limit.
method Proving a conjecture for deep networks with polynomial activation functions.
result Tight bounds on the behavior of wide networks during stochastic gradient descent and derivation of their finite-width dynamics.

Recent work by Brock et al. (2018) suggests that Generative Adversarial Networks (GANs) benefit disproportionately from large mini-batch sizes. Unfortunately, using large batches is slow and expensive on conventional hardware. Thus, it would be nice if we could generate batches that were effectively large though actual…

2019-10-29abs ↗pdf ↗

Active learning from demonstration allows a robot to query a human for specific types of input to achieve efficient learning. Existing work has explored a variety of active query strategies; however, to our knowledge, none of these strategies directly minimize the performance risk of the policy the robot is learning. U…

2019-01-08abs ↗pdf ↗

Active-GRPO improves molecular optimization by actively deciding when to imitate or self-improve.

problem Training robust and efficient molecular optimization models with large language models.
method Active-GRPO combines imitation and reinforcement learning, upgrading references and policies dynamically.
result Improves molecular optimization performance, achieving statistically significant gains.

Learning the activities of animals is important for the purpose of monitoring their welfare vis a vis their behaviour with respect to their environment and conspecifics. While previous works have largely focused on activity recognition in a single animal, little or no work has been done in learning the collective behav…

2018-11-22abs ↗pdf ↗

ZDP detects drift in large language models without labels, proving key theorems and metrics.

problem Detecting drift in large language models without task labels or output evaluations.
method Zero-Direction Probing (ZDP) framework based on null directions of transformer activations, proving theoretical guarantees.
result Proves the Variance--Leak Theorem, Fisher Null-Conservation, Rank--Leak bound, and logarithmic-regret guarantee.

Interactive machine learning improves learning efficiency with user input.

problem Expensive, time-consuming, or risky acquisition of labeled data and decision-making.
method Develops new algorithms for active learning, sequential decision making, and model selection under partial feedback.
result First efficient algorithms achieving exponential label savings and independent of action space size.

Evolutionary algorithms improve neural network performance by discovering better activation functions.

problem The choice of activation function affects neural network performance, but ReLU remains dominant.
method Defined a tree-based search space of candidate activation functions and used evolutionary algorithms (mutation, crossover, exhaustive search) to explore and discover better functions.
result Replacing ReLU with evolved activation functions statistically significantly increases network accuracy.

This paper proposes an improved active learning method using classification trees.

problem Reducing the size of training sets while maintaining high accuracy in supervised learning.
method A wrapper active learning method using a classification tree to sub-sample from low-entropy regions.
result The proposed method constructs accurate classification models even with severely restricted labeled data.

Learning automatically the best activation function for the task is an active topic in neural network research. At the moment, despite promising results, it is still difficult to determine a method for learning an activation function that is at the same time theoretically simple and easy to implement. Moreover, most of…

2019-02-08abs ↗pdf ↗

Enhanced network threat detection using KG, LLM, and imbalanced learning.

problem Challenges in network threat detection due to complex attack patterns and limited historical data.
method Integrated framework combining Knowledge Graph, Imbalanced Learning, and Large Language Model.
result Improved threat capture rate by 3%-4% and increased interpretability of risk predictions.

Active sampling improves design space exploration for analog circuits.

problem Efficiently exploring the space of design features in analog circuits with many parameters.
method Combining drastic dimension reduction with sensitivity analysis and Bayesian surrogate modeling for active sampling.
result The proposed active sampling flow outperforms traditional Monte-Carlo sampling.