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

52103155206 · Jun 202019922001200920172026
48 results for activation shifts

LCW reduces activation shift in neural networks, improving training efficiency and generalization.

problem Activation shift in neural networks leading to non-zero mean preactivation values.
method Linearly constrained weights (LCW) to reduce activation shift in fully connected and convolutional layers.
result LCW resolves the vanishing gradient problem and improves generalization of neural networks.

IUPM monitors machine learning models under gradual shifts using optimal transport and active labeling.

problem Gradual distribution shifts lead to unnoticed accuracy declines in machine learning models.
method Incremental Uncertainty-aware Performance Monitoring (IUPM) using optimal transport and active labeling.
result IUPM outperforms existing baselines in gradual shift scenarios and guides label acquisition more effectively.

The paper tackles matching a desired mean in causal systems through shift interventions.

problem Matching a desired mean in causal systems.
method Defining Markov equivalence classes, proposing active learning strategies, deriving lower bounds.
result Proposed active learning strategies require fewer interventions than previous approaches, especially for certain graph classes.

A new protocol corrects confounding effects to measure alignment-induced activation shifts accurately.

problem Confounding effects in measuring alignment-induced activation shifts using naive methods.
method Introduces a four-variant decomposition to separate alignment shift from template effects.
result Correctly measures alignment-induced activation shifts, recovering behaviorally active subspace.

New method corrects active learning for distribution shifts and outliers.

problem Conventional active learning methods fail to account for test-time distribution.
method JEPIG, a hybrid of BALD and EPIG, maximizes expected predictive information gain.
result JEPIG outperforms conventional methods in active learning with distribution shifts.

We propose a simple extension to the ReLU-family of activation functions that allows them to shift the mean activation across a layer towards zero. Combined with proper weight initialization, this alleviates the need for normalization layers. We explore the training of deep vanilla recurrent neural networks (RNNs) with…

2017-09-12abs ↗pdf ↗

CPATTA uses conformal prediction for efficient test-time adaptation.

problem Low data selection efficiency in existing ATTA methods.
method Conformal Prediction, online weight-update algorithm, domain-shift detector, staged update scheme.
result CPATTA consistently outperforms state-of-the-art methods by 5% in accuracy.

Balanced Activation improves object detection performance on long-tailed datasets.

problem Mismatch between training and testing label distributions in object detection.
method Introduces Balanced Activation (Balanced Softmax and Balanced Sigmoid) to address label distribution shift.
result Balanced Activation provides ~3% gain in mAP on LVIS-1.0 compared to state-of-the-art methods.

Discriminative active learning reduces data annotation costs for domain adaptation.

problem Conditional shift problem hinders domain adaptation between related but different domains.
method Three-stage active adversarial training: invariant feature space learning, uncertainty and diversity criteria, re-training with queried labels.
result Empirical comparisons show the proposed approach is more effective than existing methods.

We describe a mechanism by which artificial neural networks can learn rapid adaptation - the ability to adapt on the fly, with little data, to new tasks - that we call conditionally shifted neurons. We apply this mechanism in the framework of metalearning, where the aim is to replicate some of the flexibility of human …

2017-12-28abs ↗pdf ↗

Paper tackles robust classification under class-dependent domain shift.

problem Class-dependent domain shift in machine learning.
method Defined a simple optimization problem with an information theoretic constraint and solved it using neural networks.
result Demonstrated that the proposed method can learn robust classifiers that generalize well to unseen domains.

AFA evaluates AI feature acquisition strategies in domains with high costs.

problem Evaluate AI feature acquisition strategies in domains with high costs.
method Apply missing data methods and offline reinforcement learning under NDE and NUC assumptions.
result Propose a novel semi-offline reinforcement learning framework with three new estimators.

Active learning improves inspection systems by using weakly labeled data.

problem Rapidly updating machine vision inspection systems in evolving manufacturing processes.
method Developed a methodology for active learning from weakly labeled data, addressing covariate shift with domain-adversarial training.
result Demonstrated that active learning can accelerate the annotation process and reduce false positives.

Human stablecoin transactions predict political risk in cryptocurrency markets.

problem Predicting political risk in cryptocurrency markets.
method Structural break analysis and surrogate-based robustness tests.
result Human-driven stablecoin transactions shift significantly before major political events.

This paper introduces a differentiable, scalable quantization method for neural networks.

problem Previous quantization methods lacked differentiability and scalability.
method The approach is differentiable and scalable, using bit-shifting and logarithmic quantization.
result The method achieves comparable accuracy to state-of-the-art approaches with less training time and lower inference cost.

Emergent misalignment is influenced by training dynamics, model priors, and data.

problem Emergent misalignment in models
method Exploring training dynamics, model priors, and data
result Activation deltas before and after narrow fine-tuning correlate with their similarities when measured with the last prompt-token activations.

Early stopping in meta-learning improved by analyzing neural activation patterns.

problem Early stopping in few-shot learning is challenging due to distributional shifts between meta-validation and meta-test sets.
method Activation-Based Early-stopping (ABE) analyzes hidden layer activations from unlabelled support examples to detect when target generalization diverges from source data.
result Simple activation statistics can effectively estimate target generalization, improving few-shot transfer learning across various algorithms and datasets.

New regularizer improves neural network robustness and generalization.

problem Ineffective weight decay for networks with homogeneous activation functions.
method Proposes an invariant regularizer to penalize intrinsic weight norms.
result Improves generalization and adversarial robustness on various datasets.

The non-stationary nature of electroencephalography (EEG) signals makes an EEG-based brain-computer interface (BCI) a dynamic system, thus improving its performance is a challenging task. In addition, it is well-known that due to non-stationarity based covariate shifts, the input data distributions of EEG-based BCI sys…

2018-05-02abs ↗pdf ↗

Graph neural networks (GNNs) have been shown to replicate convolutional neural networks' (CNNs) superior performance in many problems involving graphs. By replacing regular convolutions with linear shift-invariant graph filters (LSI-GFs), GNNs take into account the (irregular) structure of the graph and provide meaning…

2018-10-29abs ↗pdf ↗

FreSh shifts model's initial frequency spectrum to match target signal, improving neural representation performance.

problem MLPs' low-frequency bias limits capturing high-frequency details accurately.
method FreSh selects embedding hyperparameters to align model's initial output spectrum with target signal's spectrum.
result FreSh improves performance across various neural representation methods and tasks with minimal computational overhead.

This work proposes a novel solution to the problem of internal covariate shift and dying neurons using the concept of linked neurons. We define the neuron linkage in terms of two constraints: first, all neuron activations in the linkage must have the same operating point. That is to say, all of them share input weights…

2017-12-07abs ↗pdf ↗

OSAMD adapts online to changing distributions with limited labels.

problem Models struggle with continual distribution shifts and expensive labeling in changing environments.
method Online Active Continual Adaptation with OSAMD, an online teacher-student structure and margin-based criterion.
result OSAMD achieves favorable dynamic regret bounds under changing environments with limited labels.

Model analyzes Proof-of-Stake network dynamics and speculative capital effects on token prices.

problem Understanding and managing price dynamics in Proof-of-Stake networks.
method Developed an open-economy macroeconomic model to analyze Proof-of-Stake dynamics and speculative capital effects.
result Speculative capital can shift staked-token ownership, potentially improving consensus decentralization.

New RBF networks can approximate any continuous function.

problem Approximating any continuous function on a compact subset.
method Replacing smoothing factors with shifts in RBF networks and proving approximation under certain conditions.
result RBF networks can approximate any continuous function on any compact subset.

GAPA method provides efficient uncertainty quantification for pretrained networks.

problem Reliable uncertainty estimates for pretrained models are challenging.
method Post-hoc Gaussian Process Activations (GAPA) method that shifts Bayesian modeling from weights to activations.
result GAPA method provides efficient uncertainty quantification without altering the backbone's predictions.

BOMS enhances offline MBRL by improving model selection with Bayesian optimization.

problem Inaccurate model selection in offline MBRL due to distribution shift.
method Proposes BOMS, an active model selection framework using Bayesian optimization.
result Improves model selection with only a small amount of online interaction.

We adapt Shapley values to explain model uncertainty, connecting it to information theory.

problem Explaining uncertainty in model predictions.
method Adapted Shapley value framework to quantify feature contributions to predictive uncertainty.
result Deep connections between Shapley values and information theory quantities.

Randomly initialized wide neural networks with zero-mean activations are nearly independent, potentially solving AI interpretability limits.

problem Measuring the limits of AI interpretability.
method Randomly initialized neural networks with large width and zero-mean activation functions.
result Neural networks with zero-mean activations are nearly independent, solving the computational no-coincidence conjecture.

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.

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.

Paper analyzes why deeper layers of ViTs perform worse on out-of-distribution tasks.

problem Performance degradation of intermediate layers in ViTs under distribution shift.
method Extensive linear probing experiments across various benchmarks and fine-grained analysis of transformer modules.
result Probing feedforward network activations yields best performance under significant distribution shift.

RAN model recognizes multiple activities from unlabeled sensor data.

problem Handling weakly labeled multi-activity data from wearable sensors.
method Recurrent Attention Networks (RAN) for sequential multi-activity recognition and localization.
result RAN model can infer multiple activities and determine activity locations from unlabeled data.