FiT combines transfer and meta-learning for efficient few-shot image classification.
problem Few-shot image classification in personalized and federated learning settings.
method Combines transfer learning and meta-learning with fixed pretrained backbones and fine-tuned FiLM adapter layers.
result Achieves state-of-the-art accuracy on VTAB-1k benchmark with fewer than 1% of updateable parameters.
This paper considers the problem of inferring image labels from images when only a few annotated examples are available at training time. This setup is often referred to as low-shot learning, where a standard approach is to re-train the last few layers of a convolutional neural network learned on separate classes for w…
Conventional deep learning classifiers are static in the sense that they are trained on a predefined set of classes and learning to classify a novel class typically requires re-training. In this work, we address the problem of Low-Shot network expansion learning. We introduce a learning framework which enables expandin…
Develops a method to ensure accuracy of few-shot transfer learning models.
problem Lack of generalization guarantees for low-data transfer learning.
method Trains a distribution over PEFT parameters using upstream tasks and samples plausible PEFTs for downstream tasks.
result Demonstrates non-vacuous generalization guarantees compared to existing methods in the low-shot regime.
NACs learn modular neural architectures without domain knowledge.
problem Jointly learn module configuration and execution without domain knowledge.
method Jointly trains two systems: module configuration and execution.
result Improves low-shot adaptation and OOD robustness.
The goal of this paper is to design image classification systems that, after an initial multi-task training phase, can automatically adapt to new tasks encountered at test time. We introduce a conditional neural process based approach to the multi-task classification setting for this purpose, and establish connections …
Deep neural networks suffer from over-fitting and catastrophic forgetting when trained with small data. One natural remedy for this problem is data augmentation, which has been recently shown to be effective. However, previous works either assume that intra-class variances can always be generalized to new classes, or e…
While deep learning has achieved remarkable results on various applications, it is usually data hungry and struggles to learn over non-stationary data stream. To solve these two limits, the deep learning model should not only be able to learn from a few of data, but also incrementally learn new concepts from data strea…
Despite the recent success of deep transfer learning approaches in NLP, there is a lack of quantitative studies demonstrating the gains these models offer in low-shot text classification tasks over existing paradigms. Deep transfer learning approaches such as BERT and ULMFiT demonstrate that they can beat state-of-the-…
Automatically writing stylized Chinese characters is an attractive yet challenging task due to its wide applicabilities. In this paper, we propose a novel framework named Style-Aware Variational Auto-Encoder (SA-VAE) to flexibly generate Chinese characters. Specifically, we propose to capture the different characterist…
JoLT uses LLMs to make probabilistic predictions on tabular data.
problem Making probabilistic predictions on tabular data efficiently and without preprocessing.
method JoLT leverages LLMs' in-context learning to define joint distributions over tabular data.
result JoLT outperforms other methods on tabular classification and regression tasks.
Employing deep neural networks as natural image priors to solve inverse problems either requires large amounts of data to sufficiently train expressive generative models or can succeed with no data via untrained neural networks. However, very few works have considered how to interpolate between these no- to high-data r…
This paper proposes a method to improve few-shot learning by generating multi-level weight-centric features.
problem Improving few-shot learning performance by leveraging both representation power and weight generation capacity.
method A multi-level weight-centric feature learning approach with a weight-centric training strategy and multi-level feature incorporation.
result Significantly outperforms existing methods in low-shot classification benchmarks.
ProSMIN improves representation quality through probabilistic self-supervised learning.
problem Improving representation quality in self-supervised learning.
method ProSMIN uses two neural networks, online and target, to learn diverse representations through knowledge distillation and a modified scoring rule loss function.
result ProSMIN achieves superior accuracy and calibration on various downstream tasks.
Method generates visual explanations for similarity models without classification.
problem Lack of visual explanations for similarity models trained without classification loss.
method Gradient-based visual attention using learned feature embeddings.
result Attention maps improve model performance and can be used as constraints.
GCRL learns causal factors for motion forecasting, improving out-of-distribution prediction.
problem Sensitivity to out-of-distribution data in conventional supervised learning methods.
method Generative Causal Representation Learning (GCRL) leveraging causality for knowledge transfer.
result Significantly outperforms prior models on out-of-distribution prediction.
Few-shot visual reasoning model learns analogical relationships from small data.
problem Training deep models on few samples for visual reasoning tasks.
method Meta-analogical contrastive learning to enforce structural similarity between training and test samples.
result Method outperforms state-of-the-art on RAVEN dataset with scarce training data.
Stochastic methods with coordinate-wise adaptive stepsize (such as RMSprop and Adam) have been widely used in training deep neural networks. Despite their fast convergence, they can generalize worse than stochastic gradient descent. In this paper, by revisiting the design of Adagrad, we propose to split the network par…
Sequence-to-sequence (seq2seq) based ASR systems have shown state-of-the-art performances while having clear advantages in terms of simplicity. However, comparisons are mostly done on speaker independent (SI) ASR systems, though speaker adapted conventional systems are commonly used in practice for improving robustness…
AdaPTS adapts univariate FMs for multivariate time series forecasting.
problem Challenges in managing feature dependencies and uncertainty quantification in multivariate time series forecasting.
method Adapters that transform multivariate inputs into a latent space and apply univariate FMs independently to each dimension.
result AdaPTS enhances forecasting accuracy and uncertainty quantification compared to baseline methods.
We propose a new concept named adaptive submodularity ratio to study the greedy policy for sequential decision making. While the greedy policy is known to perform well for a wide variety of adaptive stochastic optimization problems in practice, its theoretical properties have been analyzed only for a limited class of p…
Adaptive sequential decision making is one of the central challenges in machine learning and artificial intelligence. In such problems, the goal is to design an interactive policy that plans for an action to take, from a finite set of n actions, given some partial observations. It has been shown that in many applicat…
New algorithm reduces interventional strategy complexity for causal graph discovery.
problem Designing efficient interventional strategies for causal graph discovery.
method Developed an r-adaptive algorithm for causal graph discovery that minimizes the number of interventions.
result Achieved an approximation of O(min{r, log n} * n^{1/min{r, log n}}) for the verification number.
Automation of machine learning model development is increasingly becoming an established research area. While automated model selection and automated data pre-processing have been studied in depth, there is, however, a gap concerning automated model adaptation strategies when multiple strategies are available. Manually…
Paper presents a Transformer model for automatic domain adaptation.
problem Challenges in selecting or designing domain adaptation algorithms.
method Transformer model approximates and selects domain adaptation algorithms.
result Transformers can approximate and automatically select domain adaptation algorithms.
cKAM improves adaptive sampling by incorporating a cyclical stepsize scheme.
problem Adaptive Metropolis algorithms can get stuck in local modes.
method cKAM uses a cyclical stepsize scheme to encourage exploration and escape from local modes.
result cKAM successfully escapes local modes and converges to the true posterior distribution.
Adaptive variational Bayes framework improves inference adaptively.
problem Lack of general and computationally tractable variational Bayes method for adaptive inference.
method Proposes a novel adaptive variational Bayes framework combining variational posteriors over individual models.
result Adaptive variational Bayes achieves optimal contraction rates adaptively under general conditions.
New approach shows AI can adapt like toddlers by correcting old knowledge.
problem Lack of adaptability in AI models compared to humans and animals.
method Casting adaptation as posterior correction and using Bayesian Learning Rule.
result AI can learn to adapt quickly by using posterior correction.
This paper improves neural network generalization by dynamically learning kernel parameters.
problem Improving neural network generalization and adaptability.
method Diagonal adaptive kernel model that learns kernel eigenvalues and output coefficients during training.
result The diagonal adaptive kernel model significantly improves generalization over fixed-kernel methods.
Study on distributed nonparametric function estimation with optimal rate and cost of adaptation.
problem Optimal rate of convergence and cost of adaptation in distributed nonparametric function estimation.
method Distributed minimax estimation and adaptive estimation under communication constraints for Gaussian sequence model and white noise model.
result Established minimax rate of convergence and exact communication cost for adaptation.
This paper explains why Adam generalizes worse than SGD by analyzing its components.
problem Understanding why Adam generalizes worse than Stochastic Gradient Descent (SGD).
method Diffusion theoretical framework to disentangle the effects of Adaptive Learning Rate and Momentum.
result Adaptive Learning Rate helps escape saddle points but not select flat minima, while Momentum provides a drift effect to help pass through saddle points.
New adaptive importance samplers improve stability and accuracy.
problem Improving the stability and accuracy of importance sampling estimators.
method Introducing AdaOAIS, a new adaptive importance sampler using adaptive optimisers to address the instability of OAIS.
result AdaOAIS leads to stable importance sampling estimators in practice.
Adaptive networks improve model robustness through conditional normalization.
problem Limited robustness of adversarial-trained networks due to network capacity and training samples.
method Proposes a conditional normalization module to adapt networks during adversarial training.
result Adaptive networks outperform both clean validation accuracy and robustness compared to non-adaptive counterparts.
In domain adaptation, classifiers with information from a source domain adapt to generalize to a target domain. However, an adaptive classifier can perform worse than a non-adaptive classifier due to invalid assumptions, increased sensitivity to estimation errors or model misspecification. Our goal is to develop a doma…
FLAP adapts policies quickly to new tasks using shared linear representations.
problem Adapting policies to new tasks efficiently and effectively.
method FLAP uses a shared linear representation and a separate adapter network for quick adaptation.
result FLAP achieves up to 8X faster adaptation and significantly better performance on out-of-distribution tasks.
AvaGrad optimizes vision tasks by decoupling learning rate and adaptability.
problem Improving optimization methods for vision tasks.
method Derives AvaGrad, a new optimizer that decouples learning rate and adaptability.
result AvaGrad outperforms SGD on vision tasks when adaptability is properly tuned.
Learn to automatically plug domain-specific modules into a common network.
problem Learning inflexibility and computational intensiveness in multi-domain learning.
method Neural Architecture Search (NAS) for data-driven adapter plugging and structure design.
result NAS-driven MDL model achieves comparable performance to existing approaches.
FLoE adapts LLMs by selectively deploying LoRA adapters based on layer importance and task requirements.
problem Uniform LoRA deployment across all layers leads to inefficient and redundant parameter allocation.
method FLoE uses Fisher information to dynamically identify task-critical layers and optimizes LoRA ranks.
result FLoE achieves significant efficiency-accuracy trade-offs, especially in resource-constrained environments.
The paper improves generalization bounds for domain adaptation.
problem Improving generalization bounds for domain adaptation under practical conditions.
method Derives generalization bounds for domain adaptation based on finitely many moments and smoothness conditions.
result Obtains generalization bounds for domain adaptation.
New protocols show 1-bit mean estimation can be order-optimal without interaction.
problem Can 1-bit mean estimation be optimal without interaction?
method Adaptive and non-adaptive threshold and interval queries, with one adaptive transition.
result Arbitrary non-adaptive quantizers can match the adaptive rate, suggesting interaction is not necessary.
New adaptive attacks bypass many defenses to adversarial examples.
problem Adversarial example defenses are not adequately evaluated using adaptive attacks.
method Detailed analysis of thirteen defenses, demonstrating their vulnerabilities to adaptive attacks.
result Adversarial example defenses are more vulnerable to adaptive attacks than previously thought.
We study methods for aggregating pairwise comparison data in order to estimate outcome probabilities for future comparisons among a collection of n items. Working within a flexible framework that imposes only a form of strong stochastic transitivity (SST), we introduce an adaptivity index defined by the indifference se…
AdapTable adapts tabular models to shifts without source data, improving HELOC performance.
problem Distribution shifts in tabular data threaten model performance.
method Shift-aware uncertainty calibrator and label distribution handler.
result Up to 16% improvement on HELOC dataset.
The paper addresses statistical inference issues in adaptive experiments.
problem Statistical inference problems in adaptive experiments.
method Explains and fixes statistical inference issues in adaptive experiments using various methods.
result Various methods to stabilize inferences and recover asymptotic normality.
Enhances physics-informed neural networks with adaptive sampling and weighting.
problem Challenges in training physics-informed neural networks on complex problems.
method Hybrid adaptive sampling and weighting method.
result Consistently improves prediction accuracy and training efficiency.
We study the performance of the adaptive construction scheme for a Bayesian inference on the Quadratic GARCH model which introduces the asymmetry in time series dynamics. In the adaptive construction scheme a proposal density in the Metropolis-Hastings algorithm is constructed adaptively by changing the parameters of t…
We propose algorithms for online principal component analysis (PCA) and variance minimization for adaptive settings. Previous literature has focused on upper bounding the static adversarial regret, whose comparator is the optimal fixed action in hindsight. However, static regret is not an appropriate metric when the un…
New method adapts without backprop, faster and better.
problem Efficient domain adaptation without source data.
method Computing class prototypes from pre-trained model.
result Significant accuracy improvements over pre-trained model.