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.
Paper proposes CNE-net to tackle incremental learning in (T)ACSA tasks.
It was recently shown that architectural, regularization and rehearsal strategies can be used to train deep models sequentially on a number of disjoint tasks without forgetting previously acquired knowledge. However, these strategies are still unsatisfactory if the tasks are not disjoint but constitute a single increme…
DPTA improves CIL by adapting PTMs with dual prototypes.
Meta-learning approach prevents forgetting across tasks.
One of the key differences between the learning mechanism of humans and Artificial Neural Networks (ANNs) is the ability of humans to learn one task at a time. ANNs, on the other hand, can only learn multiple tasks simultaneously. Any attempts at learning new tasks incrementally cause them to completely forget about pr…
EVCL combines VCL and EWC to prevent forgetting new tasks.
A new model learns preferences incrementally without personal data.
BI-MAML learns multiple tasks without forgetting old ones.
Meta-learning approach improves object detection on new classes.
TKIL improves class-balanced performance in incremental learning.
Knowledge graph construction consists of two tasks: extracting information from external resources (knowledge population) and inferring missing information through a statistical analysis on the extracted information (knowledge completion). In many cases, insufficient external resources in the knowledge population hinde…
DoubleAdapt improves stock trend forecasting by adapting models to evolving data.
Paper presents IMRCs for evolving tasks with forward and backward learning.
Improved incremental sequence classification with temporal consistency.
FedGTEA learns new tasks in federated learning with task embeddings and alignment.
Paper proposes a method to maintain ASR performance on new tasks without forgetting old ones.
New method for incremental meta-learning reduces forgetting and improves performance.
Paper proposes a method to learn multiple tasks without forgetting, maintaining model compactness.
Proposes efficient model for continual learning that grows model over task-specific parameters.
Incremental learning suffers from two challenging problems; forgetting of old knowledge and intransigence on learning new knowledge. Prediction by the model incrementally learned with a subset of the dataset are thus uncertain and the uncertainty accumulates through the tasks by knowledge transfer. To prevent overfitti…
Paper tackles graph class-incremental learning with task profiling and prompting.
AdaAnn optimizes annealing for efficient probability density approximation.
The main purpose of incremental learning is to learn new knowledge while not forgetting the knowledge which have been learned before. At present, the main challenge in this area is the catastrophe forgetting, namely the network will lose their performance in the old tasks after training for new tasks. In this paper, we…
Transformers learn to integrate information from past positions incrementally, specializing heads in distinct patterns.
In class-incremental learning, a model learns continuously from a sequential data stream in which new classes occur. Existing methods often rely on static architectures that are manually crafted. These methods can be prone to capacity saturation because a neural network's ability to generalize to new concepts is limite…
Service robots benefit from encoding information in semantically meaningful ways to enable more robust task execution. Prior work has shown multi-relational embeddings can encode semantic knowledge graphs to promote generalizability and scalability, but only within a batched learning paradigm. We present Incremental Se…
Comprehensive surgical planning require complex patient-specific anatomical models. For instance, functional muskuloskeletal simulations necessitate all relevant structures to be segmented, which could be performed in real-time using deep neural networks given sufficient annotated samples. Such large datasets of multip…
Incremental learning from non-stationary data poses special challenges to the field of machine learning. Although new algorithms have been developed for this, assessment of results and comparison of behaviors are still open problems, mainly because evaluation metrics, adapted from more traditional tasks, can be ineffec…
Paper tackles catastrophic forgetting in neural networks with an adversarial feature alignment method.
DeGAN enriches data from related domains for future learning tasks.
This paper tackles federated incremental learning with dynamic memory allocation for improved model performance in non-IID data.
We introduce a novel sensitivity analysis framework for large scale classification problems that can be used when a small number of instances are incrementally added or removed. For quickly updating the classifier in such a situation, incremental learning algorithms have been intensively studied in the literature. Alth…
Ordinal Regression (OR) aims to model the ordering information between different data categories, which is a crucial topic in multi-label learning. An important class of approaches to OR models the problem as a linear combination of basis functions that map features to a high dimensional non-linear space. However, most…
Cross-validation (CV) is one of the main tools for performance estimation and parameter tuning in machine learning. The general recipe for computing CV estimate is to run a learning algorithm separately for each CV fold, a computationally expensive process. In this paper, we propose a new approach to reduce the computa…
Paper tackles class-incremental learning by combining self-supervised learning to mitigate prior information loss.
SpaceNet improves continual learning by intelligently compressing neural connections.
Continuous appearance shifts such as changes in weather and lighting conditions can impact the performance of deployed machine learning models. While unsupervised domain adaptation aims to address this challenge, current approaches do not utilise the continuity of the occurring shifts. In particular, many robotics appl…
PEAKS selects key training examples incrementally based on prediction error and kernel similarity.
Algorithm estimates bounds of updated classifier coefficients efficiently.
This thesis tackles continual learning in neural networks, overcoming forgetting.
Class incremental learning refers to a special multi-class classification task, in which the number of classes is not fixed but is increasing with the continual arrival of new data. Existing researches mainly focused on solving catastrophic forgetting problem in class incremental learning. To this end, however, these m…
In learning-to-learn the goal is to infer a learning algorithm that works well on a class of tasks sampled from an unknown meta distribution. In contrast to previous work on batch learning-to-learn, we consider a scenario where tasks are presented sequentially and the algorithm needs to adapt incrementally to improve i…
Resultant improves likelihood-based U-OOD detection across various tasks.
Incremental class learning, a scenario in continual learning context where classes and their training data are sequentially and disjointedly observed, challenges a problem widely known as catastrophic forgetting. In this work, we propose a novel incremental class learning method that can significantly reduce memory ove…
We introduce incremental variational inference and apply it to latent Dirichlet allocation (LDA). Incremental variational inference is inspired by incremental EM and provides an alternative to stochastic variational inference. Incremental LDA can process massive document collections, does not require to set a learning …
Paper proposes a new method to select memory data for online class-incremental learning.
Sparse Meta Networks adapt deep neural networks incrementally for fast learning.
Reinforcement learning frameworks have introduced abstractions to implement and execute algorithms at scale. They assume standardized simulator interfaces but are not concerned with identifying suitable task representations. We present Wield, a first-of-its kind system to facilitate task design for practical reinforcem…