New method reveals how training data influence diffusion model outputs.
problem Difficulty in assessing training data impact on diffusion model outputs.
method Use of ensembles trained on carefully engineered splits of training data to identify influential training examples.
result Demonstrated the viability of ensembles as generative models and validity of assessing influence.
Pre-training is crucial for learning deep neural networks. Most of existing pre-training methods train simple models (e.g., restricted Boltzmann machines) and then stack them layer by layer to form the deep structure. This layer-wise pre-training has found strong theoretical foundation and broad empirical support. Howe…
Joint training improves model accuracy by selectively using privileged information.
problem Two-stage training can lead to model failure with noisy privileged information.
method Joint training of two models to use privileged information selectively.
result Joint training outperforms two-stage baselines on synthetic and real-world tasks.
Paper proposes a new method to measure model sensitivity using final model only.
problem Understanding model behavior using only the final trained model.
method Reframe TDA as measuring sensitivity, propose further training as gold standard, unify gradient-based methods.
result Gradient-based methods approximate further training but vary in quality.
Recent work on discriminative segmental models has shown that they can achieve competitive speech recognition performance, using features based on deep neural frame classifiers. However, segmental models can be more challenging to train than standard frame-based approaches. While some segmental models have been success…
The training of Deep Neural Networks usually needs tremendous computing resources. Therefore many deep models are trained in large cluster instead of single machine or GPU. Though major researchs at present try to run whole model on all machines by using asynchronous asynchronous stochastic gradient descent (ASGD), we …
Synthetic speech data improves keyword spotting models with fewer real examples.
problem Training models for recognizing spoken keywords with limited real data.
method Used a pre-trained speech embedding model to extract features for training a small keyword spotting model.
result A model trained on synthetic speech data can detect 10 keywords with the same accuracy as a model trained on over 500 real examples.
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.
Paper tackles model vulnerabilities by reconstructing training data.
problem Reconstructing training data from model parameters poses a security risk.
method Developed a mathematical framework and score matching method for both Bayesian and non-Bayesian models.
result First score matching framework for reconstructing data in Bayesian models.
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.
New methods predict neural network quality without access to training data.
problem Predicting neural network quality without access to training or testing data.
method Meta-analysis of pretrained models using norm and power law based metrics.
result Power law based metrics can better distinguish well-trained from poorly-trained models.
RoBERTa outperforms other pre-trained models in NER tasks.
problem Improving Named Entity Recognition (NER) performance.
method Fine-tuning four pre-trained models (BERT, ERNIE, ERNIE2.0-tiny, RoBERTa) on NER task.
result RoBERTa achieved state-of-the-art results on MSRA-2006 dataset.
WAFFLE embeds watermarks in federated learning models without access to training data.
problem Model theft in federated learning due to lack of centralized control.
method Retraining step at server after model aggregation, generating backdoor as watermark.
result Efficient embedding of resilient watermark with negligible accuracy loss.
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.
SMART training improves mask-predict translations.
problem Closing the performance gap between semi-autoregressive and autoregressive models.
method SMART training method for conditional masked language models.
result SMART-trained models produce higher-quality translations.
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.
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.
SMT trains generative models by estimating mixture scores, outperforming existing methods.
problem Training one-step generative models efficiently and effectively.
method Score-of-Mixture Training (SMT) estimates the score of mixture distributions between real and fake samples.
result SMT/SMD outperform existing methods on CIFAR-10 and ImageNet 64x64 datasets.
Survey on reducing deep neural network training complexity.
problem Efficient training of large deep neural networks.
method Pruning and freezing parts of the network during training.
result Dimensionality reduction improves training efficiency.
Predicts optimal training dataset sizes per class for machine learning models.
problem Optimizing training dataset sizes for class-specific machine learning models.
method Algorithm based on space-filling design of experiments, models like powerlaw curves and generalized linear models.
result The algorithm predicts optimal training dataset sizes per class for improved model performance.
New method improves neural spike train models by minimizing divergence directly, leading to better performance.
problem Poor performance and divergence issues in spike train models using maximum likelihood estimation.
method Directly minimize maximum mean discrepancy using spike train kernels and stochastic optimization.
result The proposed method generates well-behaved models with better control over feature trade-offs.
We use GG distributions to optimize LLMs, reducing size and training time.
problem Lack of understanding in the statistical structure of LLMs.
method GG-based initialization, ACT, GCT.
result Smaller, faster models with minimal communication overhead.
Machine learning models are vulnerable to adversarial examples. Iterative adversarial training has shown promising results against strong white-box attacks. However, adversarial training is very expensive, and every time a model needs to be protected, such expensive training scheme needs to be performed. In this paper,…
Gradient amplification boosts deep learning model performance without increasing training time.
problem Vanishing gradients in deep neural networks.
method Gradient amplification approach to prevent vanishing gradients and training strategy to enable/disable across epochs.
result Improves performance of deep learning models with reduced training time.
MGD improves ML training efficiency and accuracy.
problem Optimizing large-scale ML model training configurations.
method Metagradient descent and smooth model training.
result Significantly improved model performance and learning rate schedules.
Model trains passing events on a bridge using multilevel Gaussian process.
problem Represent aggregate train-passing events from a bridge monitoring system.
method Formulate a combined model with low-rank approximation hierarchical Gaussian process, incorporating domain expertise as constraints.
result Allow for simulation of previously unobserved train types.
Adversarial samples are perturbed inputs crafted to mislead the machine learning systems. A training mechanism, called adversarial training, which presents adversarial samples along with clean samples has been introduced to learn robust models. In order to scale adversarial training for large datasets, these perturbati…
The paper studies stability of generative models trained on mixed data.
problem Training generative models on mixed datasets (real and synthetic data).
method Developed a framework to rigorously study stability under specific conditions.
result Proved the stability of iterative training under certain conditions.
Theoretical study shows adversarial training improves robustness in deep learning models.
problem Ensuring robustness in pre-trained deep learning models.
method Theoretical analysis of adversarial training and feature purification in two-layer neural networks.
result Adversarial training leads to feature purification, making models more robust to attacks.
Auxiliary Tuning adapts pre-trained models for novel tasks efficiently.
problem Adapting pre-trained models for new tasks efficiently.
method Supplementing pre-trained model with an auxiliary model that shifts output distribution.
result Achieved similar results to training from scratch with fewer resources.
LEMON uses pre-trained models to scale neural networks efficiently.
problem Efficiency in scaling deep neural networks, especially Transformers, which are resource-intensive to train from scratch.
method LEMON initializes scaled models using pre-trained weights and optimizes learning rates.
result Significant reduction in training time and computational costs for Vision Transformers and BERT.
MACER trains robust models without adversarial training, faster and more effective.
problem Learning robust models without relying on attack-dependent adversarial training.
method MACER trains provably robust smoothed classifiers by maximizing certified radius.
result MACER achieves larger average certified radius and faster training time compared to state-of-the-art methods.
Deep convolutional neural networks have achieved great success in various applications. However, training an effective DNN model for a specific task is rather challenging because it requires a prior knowledge or experience to design the network architecture, repeated trial-and-error process to tune the parameters, and …
In this work, three lattice-free (LF) discriminative training criteria for purely sequence-trained neural network acoustic models are compared on LVCSR tasks, namely maximum mutual information (MMI), boosted maximum mutual information (bMMI) and state-level minimum Bayes risk (sMBR). We demonstrate that, analogous to L…
For many large undirected models that arise in real-world applications, exact maximumlikelihood training is intractable, because it requires computing marginal distributions of the model. Conditional training is even more difficult, because the partition function depends not only on the parameters, but also on the obse…
End-to-end training of automated speech recognition (ASR) systems requires massive data and compute resources. We explore transfer learning based on model adaptation as an approach for training ASR models under constrained GPU memory, throughput and training data. We conduct several systematic experiments adapting a Wa…
Adversarially-trained models transfer better in limited data scenarios.
problem Improving transfer learning performance with limited data.
method Adversarially-train deep nets, freeze early layers, fine-tune last layers, observe shape vs texture bias.
result Adversarially-trained models transfer better than non-adversarially-trained models, especially with limited data.
Pre-trained model from healthy ADLs improves gait pattern classification for Parkinson's disease.
problem Limited training data for deep learning models in healthcare applications.
method Used convolutional autoencoder to extract features from healthy ADLs data and trained a multi-layer perceptron model for Parkinson's disease classification.
result Features extracted from healthy ADLs data can be used to train an effective classification model for Parkinson's disease.
Develops a fast method to analyze image similarities without pre-trained models.
problem Lack of efficient methods to analyze fine-level similarities in large image datasets.
method Wavelet decomposition and numerical analysis tools.
result Identifies similar images in standard datasets (like CIFAR) in a few seconds.
This study explains RL training dynamics in LLMs, focusing on token-level optimization and reasoning pattern reshaping.
problem Understanding the training dynamics of RL in LLMs to improve their reasoning capabilities.
method Empirical analysis and theoretical modeling of RL training process, focusing on reasoning patterns and token optimization.
result RL primarily optimizes a sparse subset of critical tokens, reshaping reasoning pattern distributions and affecting model performance.
This paper proposes a model of information cascades as directed spanning trees (DSTs) over observed documents. In addition, we propose a contrastive training procedure that exploits partial temporal ordering of node infections in lieu of labeled training links. This combination of model and unsupervised training makes …
The paper analyzes how adding gradient data affects overparameterized models' performance.
problem The impact of Sobolev training on high-dimensional predictive models.
method Combining replica method from statistical physics and operator-valued free probability theory.
result Sobolev training does not universally improve performance for target functions described by single-index models.
Paper tackles model collapse in recursive generative models using a weighted training scheme.
problem Model collapse in recursive generative models trained on synthetic data.
method Iteratively trains models on real and synthetic data, evaluates weighted training schemes.
result Optimal weighting scheme for synthetic data follows a unified expression, revealing a trade-off with model performance.
Self-supervised skip-tree training improves mathematical reasoning in language models.
problem Improving logical reasoning in language models for formal mathematics.
method Self-supervised language modeling on mathematical formulas, skip-tree task.
result Models trained on skip-tree task outperform standard models in mathematical reasoning tasks.
Proposes a method to update model weights by dynamically changing features during training.
problem Updating model weights efficiently during training.
method Dynamic feature replacement during training to update model weights.
result Increased model accuracy with same training time and complexity.
Surrogate models speed up RL training in dynamic systems.
problem High computational cost of high-fidelity simulations.
method Developed and tested surrogate models for RL training.
result Surrogate models can significantly accelerate RL training.
Adversarial training was introduced as a way to improve the robustness of deep learning models to adversarial attacks. This training method improves robustness against adversarial attacks, but increases the models vulnerability to privacy attacks. In this work we demonstrate how model inversion attacks, extracting trai…
A novel solve-training framework is proposed to train neural network in representing low dimensional solution maps of physical models. Solve-training framework uses the neural network as the ansatz of the solution map and train the network variationally via loss functions from the underlying physical models. Solve-trai…