New method improves ensemble quality by exploring the pre-train basin more effectively.
problem Limited diversity in ensembles trained from a single pre-trained checkpoint.
method Proposed StarSSE modification of Snapshot Ensembles for transfer learning.
result Stronger ensembles and uniform model soups achieved.
Ensemble pre-trained models for low data transfer learning.
problem Training good models from scratch in low data regime.
method Fine-tune nearest-neighbour ranked pre-trained models to create ensembles.
result Achieves state-of-the-art performance with lower inference budget.
Robust cancer screening model using pre-trained ensembles for biomarkers.
problem Detecting early-stage cancer, especially in hard-to-diagnose cases like pancreatic cancer.
method Meta-trained Hyperfast model for robust classification, combined with ensembling of XGBoost and LightGBM.
result Achieved highest AUC of 0.9929 and robust performance on imbalanced datasets.
Deep Neural Networks (DNNs) have become increasingly popular in computer vision, natural language processing, and other areas. However, training and fine-tuning a deep learning model is computationally intensive and time-consuming. We propose a new method to improve the performance of nearly every model including pre-t…
We present local ensembles, a method for detecting underspecification -- when many possible predictors are consistent with the training data and model class -- at test time in a pre-trained model. Our method uses local second-order information to approximate the variance of predictions across an ensemble of models from…
Revises Bayesian model averaging for foundation models.
problem Ensemble pre-trained and lightly-finetuned foundation models for improved classification performance.
method Introduces trainable linear classifiers and computationally cheaper model averaging scheme (OMA).
result Ensembled models can better predict on various datasets.
Differentially private ensemble classifiers adapt to data streams while protecting privacy.
problem Adapting to evolving data characteristics while protecting private information.
method Unbounded ensemble updates, model agnostic approach.
result Outperforms competitors on various privacy, drift, and distribution settings.
PEP improves deep network performance and calibration by perturbing optimal parameters.
problem Improving deep network performance and calibration.
method Parameter Ensembling by Perturbation (PEP) constructs an ensemble of parameter values as random perturbations of the optimal set, maximizing log-likelihood on validation data.
result PEP provides a small to substantial improvement in calibration and log-likelihood, and in some cases, classification accuracy.
A new method aggregates generative classifiers to resist adversarial attacks.
problem Adversarial attacks on deep neural networks.
method Rank-aggregating ensemble of generative classifiers trained on intermediate layer responses.
result The ensemble of generative classifiers shows robustness to adversarial attacks.
EnKG solves inverse problems without derivatives, using diffusion models.
problem Solving inverse problems with derivative-free methods.
method Ensemble Kalman Diffusion Guidance (EnKG) using diffusion models.
result EnKG can solve inverse problems with only forward model evaluations.
Training a neural network model can be a lifelong learning process and is a computationally intensive one. A severe adverse effect that may occur in deep neural network models is that they can suffer from catastrophic forgetting during retraining on new data. To avoid such disruptions in the continuous learning, one ap…
New method uses diffusion models for inverse problems without approximations.
problem Solving complex inverse problems in high dimensions.
method Ensemble-based algorithm using diffusion models without approximations.
result Empirically validated method gives more accurate reconstructions.
Voting ensemble of robust models improves robustness.
problem Improving robustness of defensive models against adversarial attacks.
method Hard-label voting ensemble of pretrained robust models.
result Voting ensemble can boost robust error over individual models.
Paper discovers simplicial complexes connecting trained models for improved ensembling.
problem Improving robustness and accuracy of deep learning ensembles.
method Identifies mode-connecting simplicial complexes on loss surfaces.
result Efficiently builds simplicial complexes for ensembling, outperforming independent ensembles.
In this study, a novel sparsity-driven weighted ensemble classifier (SDWEC) that improves classification accuracy and minimizes the number of classifiers is proposed. Using pre-trained classifiers, an ensemble in which base classifiers votes according to assigned weights is formed. These assigned weights directly affec…
Transformer pre-training improves stock return prediction accuracy.
problem Improving stock price prediction accuracy for better investment decisions.
method Pre-trained transformer models on TSX index, fine-tuned for individual stocks, compared to LSTM and XGBoost.
result Transformer model achieved lower mean squared error than benchmarks.
Transformer models improve malware classification, especially for imbalanced datasets.
problem Imbalanced multiclass malware classification.
method Bagging-based random transformer forest (RTF) ensemble of pre-trained transformer models (BERT or CANINE).
result Ensemble of pre-trained transformer models achieves state-of-the-art F1-score of 0.6149 on a benchmark dataset.
Capturing sentence semantics plays a vital role in a range of text mining applications. Despite continuous efforts on the development of related datasets and models in the general domain, both datasets and models are limited in biomedical and clinical domains. The BioCreative/OHNLP organizers have made the first attemp…
Combines Laplace approximations of deep networks for better uncertainty quantification.
problem Overconfident predictions on outliers in deep learning models.
method Gaussian mixture model posterior using weighted sum of Laplace approximations of pre-trained deep networks.
result Mitigates overconfidence 'far away' from training data.
Compact Gaussian model approximates deep ensemble predictions.
problem Efficiently approximating deep ensemble models for image prediction.
method Sparse-structured multivariate Gaussian with Cholesky parameterization trained to match pre-trained ensemble outputs.
result Compact representation captures uncertainty and structured correlations explicitly.
ABNN converts pre-trained DNNs into BNNs for reliable uncertainty quantification.
problem Uncertainty quantification in deep neural networks (DNNs) is challenging and critical for real-world applications.
method Adaptable Bayesian Neural Network (ABNN) that transforms pre-trained DNNs into BNNs with minimal overhead.
result ABNN achieves state-of-the-art performance in image classification and semantic segmentation tasks.
Automates zero-shot classification by scoring and weighting prompts.
problem Improving zero-shot accuracy through prompt ensembling.
method Automatic prompt scoring and weighting method.
result Method outperforms existing techniques on various benchmarks.
POETS optimizes LLMs by combining policy ensembles and KL regularization.
problem Balancing exploration and exploitation in decision-making and optimization.
method POETS uses policy ensembles and KL regularization to optimize LLMs efficiently.
result POETS achieves state-of-the-art sample efficiency across various domains.
Reinforcement learning (RL) methods have been shown to be capable of learning intelligent behavior in rich domains. However, this has largely been done in simulated domains without adequate focus on the process of building the simulator. In this paper, we consider a setting where we have access to an ensemble of pre-tr…
In this appraisal paper, we evaluate the efficacy of SHIELD, a compression-based defense framework for countering adversarial attacks on image classification models, which was published at KDD 2018. Here, we consider alternative threat models not studied in the original work, where we assume that an adaptive adversary …
We propose a Bayesian evidence framework to facilitate transfer learning from pre-trained deep convolutional neural networks (CNNs). Our framework is formulated on top of a least squares SVM (LS-SVM) classifier, which is simple and fast in both training and testing, and achieves competitive performance in practice. The…
Large-scale datasets may contain significant proportions of noisy (incorrect) class labels, and it is well-known that modern deep neural networks (DNNs) poorly generalize from such noisy training datasets. To mitigate the issue, we propose a novel inference method, termed Robust Generative classifier (RoG), applicable …
In this paper we describe our attempt at producing a state-of-the-art Twitter sentiment classifier using Convolutional Neural Networks (CNNs) and Long Short Term Memory (LSTMs) networks. Our system leverages a large amount of unlabeled data to pre-train word embeddings. We then use a subset of the unlabeled data to fin…
GP-TS optimizes TLM pre-training hyperparameters efficiently.
problem Resource inefficiency in TLM pre-training.
method Bayesian optimization with Thompson sampling and Gaussian process.
result GP-TS achieves lower MLM loss in fewer epochs.
Self-training outperforms pre-training on COCO object detection and segmentation datasets.
problem The effectiveness of pre-training in improving object detection and segmentation models is limited.
method Investigated self-training as an alternative method to utilize additional data.
result Self-training consistently improves model performance across various dataset sizes and data augmentation levels.
Developing effective and efficient recommendation methods is very challenging for modern e-commerce platforms. Generally speaking, two essential modules named "Click-Through Rate Prediction" (\textit{CTR}) and "Conversion Rate Prediction" (\textit{CVR}) are included, where \textit{CVR} module is a crucial factor that a…
He et al. (2018) have called into question the utility of pre-training by showing that training from scratch can often yield similar performance to pre-training. We show that although pre-training may not improve performance on traditional classification metrics, it improves model robustness and uncertainty estimates. …
SAFER method certifies robustness to word substitutions without model structure.
problem Certified robustness against synonymous word substitutions in NLP models.
method Randomized smoothing with stochastic ensemble of randomized inputs.
result Significantly outperforms state-of-the-art methods for certified robustness.
New framework explains how larger pre-trained models reduce downstream learning sample complexity.
problem Understanding why larger pre-trained models reduce sample complexity in downstream tasks.
method Introducing a novel framework called Caulking inspired by PEFT methods.
result Improved pre-trained models provably decrease downstream task sample complexity.
Pre-training on different modalities improves Transformer performance in offline reinforcement learning.
problem Improving Transformer-based models for offline reinforcement learning tasks.
method Empirical investigation of pre-training on language and vision data, analysis of internal representations, and fine-tuning with no context.
result Pre-trained Transformers with language data can utilize context-like information to solve reinforcement learning tasks efficiently.
The study finds a theoretical bound for pre-training iterations needed for pruning to yield good subnetwork performance.
problem Discovering efficient subnetworks within pre-trained dense networks.
method Mathematical analysis of a two-layer, fully-connected network, validating with a multi-layer perceptron trained on MNIST.
result A logarithmically dependent threshold on dataset size for successful pruning.
Unsupervised pre-training improves model generalization, but lacks theoretical understanding.
problem Lack of theoretical understanding of unsupervised pre-training's impact on model generalization.
method Introduces a novel theoretical framework to analyze and enhance generalization.
result Enhances understanding of unsupervised pre-training and fine-tuning, proposing a new regularization method.
Paper shows pre-training and transfer learning reduce sample complexity for neural networks.
problem Training high-dimensional supervised learning with limited labeled data.
method Study of single-layer neural networks via online stochastic gradient descent, considering concept shift.
result Pre-training and transfer learning reduce sample complexity by polynomial factors under general assumptions.
Fine-tuning with pre-training data improves performance.
problem Limited training data for tasks.
method Theoretical analysis of excess risk bound and selection of pre-training data subset.
result Improvement in generalization performance with pre-training data.
This paper connects masked pre-training to Bayesian model selection.
problem Understanding the success of masked pre-training and its generalization.
method The paper shows masked pre-training corresponds to maximizing the marginal likelihood.
result Masked pre-training with a suitable scoring function maximizes the marginal likelihood.
U-Cast simplifies AI weather forecasting with a standard U-Net and efficient training.
problem Complex AI models limit accessibility and cost for weather forecasting.
method Simple U-Net backbone, deterministic pre-training, and probabilistic fine-tuning with Monte Carlo Dropout.
result U-Cast matches or exceeds state-of-the-art models in accuracy while reducing training and inference costs.
New attack reveals memorization patterns in pre-trained LLMs.
problem Determining if a data point was part of a pre-trained LLM's training set.
method Adapts MIA statistical tests to LLM's perplexity dynamics of subsequences.
result Significantly outperforms prior approaches in membership inference attacks.
Named Entity Recognition (NER) is a fundamental Natural Language Processing (NLP) task to extract entities from unstructured data. The previous methods for NER were based on machine learning or deep learning. Recently, pre-training models have significantly improved performance on multiple NLP tasks. In this paper, fir…
This work studies the impact of intra-/inter-class diversity on pre-training datasets and finds a balance for optimal performance.
problem The impact of intra-/inter-class diversity on supervised pre-training datasets and their effect on downstream tasks.
method Empirical study and theoretical analysis of the relationship between diversity types and downstream performance.
result The optimal class-to-sample ratio is invariant to the size of the pre-training dataset and can be predicted.
Paper develops a theory explaining contrastive pre-training for multimodal AI.
problem Limited theoretical understanding of contrastive pre-training for multi-modal AI.
method Introduces approximate sufficient statistics and Joint Generative Hierarchical Model.
result Near-minimizers of contrastive loss are approximately sufficient, enabling diverse downstream tasks.
The paper develops an asymptotic theory of self-supervised pre-training.
problem Sharpness of current rates in self-supervised pre-training and their accuracy.
method Two-stage M-estimation and tools from Riemannian geometry.
result Characterization of the limiting distribution of the downstream test risk.
Bayesian neural networks with functional priors improve surrogate modeling in mechanics.
problem Challenges in integrating prior knowledge and quantifying uncertainties in high-dimensional NN parameter spaces.
method Anchored ensembling to integrate a priori information and learn low-rank correlations between NN parameters.
result Effective transfer of knowledge between function-space and parameter-space priors improves surrogate model accuracy and uncertainty estimation.
New analysis shows diverse classes in pre-training boost NLP performance.
problem Improving sample efficiency in downstream NLP tasks.
method Proved that diverse classes in pre-training lead to better performance, using a large last linear layer singular value.
result Transfer learning excess risk improves with large ildeν and $O\left(\frac{1}{ ildeν \sqrt{n}}
ight)$ rate.