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.
Bayesian MAML outperforms MAML in meta learning tasks with theoretical guarantees.
BayPrAnoMeta tackles few-shot industrial image anomaly detection with Bayesian methods.
New PAC-Bayesian bounds explain few-shot learning performance gaps.
Sharp-MAML improves MAML by reducing saddle points in few-shot learning.
We derive the ODE of MAML and propose a new BI-MAML algorithm.
We study the convergence of a class of gradient-based Model-Agnostic Meta-Learning (MAML) methods and characterize their overall complexity as well as their best achievable accuracy in terms of gradient norm for nonconvex loss functions. We start with the MAML method and its first-order approximation (FO-MAML) and high…
We introduce ES-MAML, a new framework for solving the model agnostic meta learning (MAML) problem based on Evolution Strategies (ES). Existing algorithms for MAML are based on policy gradients, and incur significant difficulties when attempting to estimate second derivatives using backpropagation on stochastic policies…
The study examines when MAML's objective has a benign landscape.
MAML outperforms NAL in diverse task landscapes.
Model-agnostic meta-learning (MAML) is a meta-learning technique to train a model on a multitude of learning tasks in a way that primes the model for few-shot learning of new tasks. The MAML algorithm performs well on few-shot learning problems in classification, regression, and fine-tuning of policy gradients in reinf…
The field of few-shot learning has recently seen substantial advancements. Most of these advancements came from casting few-shot learning as a meta-learning problem. Model Agnostic Meta Learning or MAML is currently one of the best approaches for few-shot learning via meta-learning. MAML is simple, elegant and very pow…
MAML with over-parameterized DNNs converges globally at a linear rate.
An important research direction in machine learning has centered around developing meta-learning algorithms to tackle few-shot learning. An especially successful algorithm has been Model Agnostic Meta-Learning (MAML), a method that consists of two optimization loops, with the outer loop finding a meta-initialization, f…
New framework guarantees convergence of multi-step MAML.
BI-MAML learns multiple tasks without forgetting old ones.
ST-MAML tackles task ambiguity in meta-learning by encoding tasks with stochastic representations.
Efficiently adapting to new environments and changes in dynamics is critical for agents to successfully operate in the real world. Reinforcement learning (RL) based approaches typically rely on external reward feedback for adaptation. However, in many scenarios this reward signal might not be readily available for the …
La-MAML improves fast online continual learning with a look-ahead approach.
MAML optimizes shared priors for subtasks in a nonconvex meta-objective.
Meta-learning techniques like MAML and Reptile fail to generalize well to out-of-distribution tasks compared to simple finetuning.
A new meta-learning framework that assigns weights to source tasks based on target samples.
New algorithm TURTLE outperforms MAML and meta-learner LSTM.
Study MAML's generalization in varying tasks, proving bounds on error.
The paper analyzes MAML's representation using RSA, revealing that feature reuse is not the primary reason for its success.
This work is a reproducibility study of the paper of Antoniou and Storkey [2019], published at NeurIPS 2019. Our results are in parts similar to the ones reported in the original paper, supporting the central claim of the paper that the proposed novel method, called Self-Critique and Adapt (SCA), improves the performan…
New analysis improves generalization bounds for meta-learning.
Unified proof for scalable personalized federated learning.
Model-Agnostic Meta-Learning (MAML) and its variants have achieved success in meta-learning tasks on many datasets and settings. On the other hand, we have just started to understand and analyze how they are able to adapt fast to new tasks. For example, one popular hypothesis is that the algorithms learn good represent…
Bayesian online meta-learning framework tackles catastrophic forgetting in few-shot classification.
New method improves convergence of RL meta-learning.
Data sets for fairness relevant tasks can lack examples or be biased according to a specific label in a sensitive attribute. We demonstrate the usefulness of weight based meta-learning approaches in such situations. For models that can be trained through gradient descent, we demonstrate that there are some parameter co…
VMGP extends Gaussian processes for Bayesian meta-learning, improving uncertainty prediction.
Gradient-based meta-learners such as MAML are able to learn a meta-prior from similar tasks to adapt to novel tasks from the same distribution with few gradient updates. One important limitation of such frameworks is that they seek a common initialization shared across the entire task distribution, substantially limiti…
Meta-learning has emerged as an important framework for learning new tasks from just a few examples. The success of any meta-learning model depends on (i) its fast adaptation to new tasks, as well as (ii) having a shared representation across similar tasks. Here we extend the model-agnostic meta-learning (MAML) framewo…
BOIL updates model body only, showing better few-shot learning performance.
Current deep learning based text classification methods are limited by their ability to achieve fast learning and generalization when the data is scarce. We address this problem by integrating a meta-learning procedure that uses the knowledge learned across many tasks as an inductive bias towards better natural languag…
Bayesian meta-learning on relation graphs improves few-shot relation extraction.
We propose CAVIA for meta-learning, a simple extension to MAML that is less prone to meta-overfitting, easier to parallelise, and more interpretable. CAVIA partitions the model parameters into two parts: context parameters that serve as additional input to the model and are adapted on individual tasks, and shared param…
Fine-tuning improves meta-learning by leveraging shared representations.
Federated Learning (FL) refers to learning a high quality global model based on decentralized data storage, without ever copying the raw data. A natural scenario arises with data created on mobile phones by the activity of their users. Given the typical data heterogeneity in such situations, it is natural to ask how ca…
New method generates universal adversarial perturbations across different image sources.
A core capability of intelligent systems is the ability to quickly learn new tasks by drawing on prior experience. Gradient (or optimization) based meta-learning has recently emerged as an effective approach for few-shot learning. In this formulation, meta-parameters are learned in the outer loop, while task-specific m…
ANIL adapts only a subset of parameters, reducing computational cost.
We propose meta-curvature (MC), a framework to learn curvature information for better generalization and fast model adaptation. MC expands on the model-agnostic meta-learner (MAML) by learning to transform the gradients in the inner optimization such that the transformed gradients achieve better generalization performa…
Method improves few-shot one-class classification.
Fully automating machine learning pipelines is one of the key challenges of current artificial intelligence research, since practical machine learning often requires costly and time-consuming human-powered processes such as model design, algorithm development, and hyperparameter tuning. In this paper, we verify that au…
Meta-reinforcement learning improves fault-adaptive control efficiency.
Many real-world problems, including multi-speaker text-to-speech synthesis, can greatly benefit from the ability to meta-learn large models with only a few task-specific components. Updating only these task-specific modules then allows the model to be adapted to low-data tasks for as many steps as necessary without ris…