A Policy for Early Sequence Classification

• URL: http://arxiv.org/abs/2304.03463v1
• Date: Fri, 7 Apr 2023 03:38:54 GMT
• Title: A Policy for Early Sequence Classification
• Authors: Alexander Cao, Jean Utke and Diego Klabjan
• Abstract summary: We introduce a novel method to classify a sequence as soon as possible without waiting for the last element. Our method achieves an average AUC increase of 11.8% over multiple experiments.
• Score: 86.80932013694684
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Sequences are often not received in their entirety at once, but instead, received incrementally over time, element by element. Early predictions yielding a higher benefit, one aims to classify a sequence as accurately as possible, as soon as possible, without having to wait for the last element. For this early sequence classification, we introduce our novel classifier-induced stopping. While previous methods depend on exploration during training to learn when to stop and classify, ours is a more direct, supervised approach. Our classifier-induced stopping achieves an average Pareto frontier AUC increase of 11.8% over multiple experiments.

Related papers

• Almost Sure Convergence of Average Reward Temporal Difference Learning [20.474661995490365]
  Tabular average reward Temporal (TD) learning is perhaps the simplest and the most fundamental policy evaluation algorithm. We are the first to prove that, under very mild conditions, average reward TD converges almost surely to a sample path dependent fixed point. Key to this success is a new general Difference approximation result concerning nonexpansive mappings with Markovian and additive noise.
  arXiv  Detail & Related papers  (2024-09-29T04:16:24Z)
• SLCA++: Unleash the Power of Sequential Fine-tuning for Continual Learning with Pre-training [68.7896349660824]
  We present an in-depth analysis of the progressive overfitting problem from the lens of Seq FT. Considering that the overly fast representation learning and the biased classification layer constitute this particular problem, we introduce the advanced Slow Learner with Alignment (S++) framework. Our approach involves a Slow Learner to selectively reduce the learning rate of backbone parameters, and a Alignment to align the disjoint classification layers in a post-hoc fashion.
  arXiv  Detail & Related papers  (2024-08-15T17:50:07Z)
• Online conformal prediction with decaying step sizes [15.884682750072399]
  We introduce a method for online conformal prediction with decaying step sizes. Unlike previous methods, we can simultaneously estimate a population quantile when it exists.
  arXiv  Detail & Related papers  (2024-02-02T04:42:09Z)
• FeCAM: Exploiting the Heterogeneity of Class Distributions in Exemplar-Free Continual Learning [21.088762527081883]
  Exemplar-free class-incremental learning (CIL) poses several challenges since it prohibits the rehearsal of data from previous tasks. Recent approaches to incrementally learning the classifier by freezing the feature extractor after the first task have gained much attention. We explore prototypical networks for CIL, which generate new class prototypes using the frozen feature extractor and classify the features based on the Euclidean distance to the prototypes.
  arXiv  Detail & Related papers  (2023-09-25T11:54:33Z)
• Early Classifying Multimodal Sequences [86.80932013694684]
  Trading wait time for decision certainty leads to early classification problems. We show our new method yields experimental AUC advantages of up to 8.7%.
  arXiv  Detail & Related papers  (2023-05-02T01:57:34Z)
• Large-scale Pre-trained Models are Surprisingly Strong in Incremental Novel Class Discovery [76.63807209414789]
  We challenge the status quo in class-iNCD and propose a learning paradigm where class discovery occurs continuously and truly unsupervisedly. We propose simple baselines, composed of a frozen PTM backbone and a learnable linear classifier, that are not only simple to implement but also resilient under longer learning scenarios.
  arXiv  Detail & Related papers  (2023-03-28T13:47:16Z)
• Positive-Unlabeled Classification under Class-Prior Shift: A Prior-invariant Approach Based on Density Ratio Estimation [85.75352990739154]
  We propose a novel PU classification method based on density ratio estimation. A notable advantage of our proposed method is that it does not require the class-priors in the training phase.
  arXiv  Detail & Related papers  (2021-07-11T13:36:53Z)
• Class-incremental Learning using a Sequence of Partial Implicitly Regularized Classifiers [0.0]
  In class-incremental learning, the objective is to learn a number of classes sequentially without having access to the whole training data. Our experiments on CIFAR100 dataset show that the proposed method improves the performance over SOTA by a large margin.
  arXiv  Detail & Related papers  (2021-04-04T10:02:45Z)
• PANDA: Adapting Pretrained Features for Anomaly Detection and Segmentation [34.98371632913735]
  We show that combining pretrained features with simple anomaly detection and segmentation methods convincingly outperforms state-of-the-art methods. In order to obtain further performance gains, we adapt pretrained features to the target distribution.
  arXiv  Detail & Related papers  (2020-10-12T17:52:50Z)
• Predicting MOOCs Dropout Using Only Two Easily Obtainable Features from the First Week's Activities [56.1344233010643]
  Several features are considered to contribute towards learner attrition or lack of interest, which may lead to disengagement or total dropout. This study aims to predict dropout early-on, from the first week, by comparing several machine-learning approaches.
  arXiv  Detail & Related papers  (2020-08-12T10:44:49Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.