Semantic Curiosity for Active Visual Learning

• URL: http://arxiv.org/abs/2006.09367v1
• Date: Tue, 16 Jun 2020 17:59:24 GMT
• Title: Semantic Curiosity for Active Visual Learning
• Authors: Devendra Singh Chaplot, Helen Jiang, Saurabh Gupta, Abhinav Gupta
• Abstract summary: We study the task of embodied interactive learning for object detection. Our goal is to learn an object detector by having an agent select what data to obtain labels for.
• Score: 45.75355448193764
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: In this paper, we study the task of embodied interactive learning for object detection. Given a set of environments (and some labeling budget), our goal is to learn an object detector by having an agent select what data to obtain labels for. How should an exploration policy decide which trajectory should be labeled? One possibility is to use a trained object detector's failure cases as an external reward. However, this will require labeling millions of frames required for training RL policies, which is infeasible. Instead, we explore a self-supervised approach for training our exploration policy by introducing a notion of semantic curiosity. Our semantic curiosity policy is based on a simple observation -- the detection outputs should be consistent. Therefore, our semantic curiosity rewards trajectories with inconsistent labeling behavior and encourages the exploration policy to explore such areas. The exploration policy trained via semantic curiosity generalizes to novel scenes and helps train an object detector that outperforms baselines trained with other possible alternatives such as random exploration, prediction-error curiosity, and coverage-maximizing exploration.

Related papers

• Learning to Explore Informative Trajectories and Samples for Embodied Perception [24.006056116516618]
  Generalizing perception models to unseen embodied tasks is insufficiently studied. We build a 3D semantic distribution map to train the exploration policy self-supervised. With the explored informative trajectories, we propose to select hard samples on trajectories based on the semantic distribution uncertainty. Experiments show that the perception model fine-tuned with our method outperforms the baselines trained with other exploration policies.
  arXiv  Detail & Related papers  (2023-03-20T08:20:04Z)
• Look Around and Learn: Self-Training Object Detection by Exploration [23.620820805804616]
  An agent learns to explore the environment using a pre-trained off-the-shelf detector to locate objects and associate pseudo-labels. By assuming that pseudo-labels for the same object must be consistent across different views, we learn the exploration policy Look Around to mine hard samples. We implement a unified benchmark of the current state-of-the-art and compare our approach with pre-existing exploration policies and perception mechanisms.
  arXiv  Detail & Related papers  (2023-02-07T16:26:45Z)
• Semantic Novelty Detection via Relational Reasoning [17.660958043781154]
  We propose a novel representation learning paradigm based on relational reasoning. Our experiments show that this knowledge is directly transferable to a wide range of scenarios. It can be exploited as a plug-and-play module to convert closed-set recognition models into reliable open-set ones.
  arXiv  Detail & Related papers  (2022-07-18T15:49:27Z)
• SEAL: Self-supervised Embodied Active Learning using Exploration and 3D Consistency [122.18108118190334]
  We present a framework called Self- Embodied Embodied Active Learning (SEAL) It utilizes perception models trained on internet images to learn an active exploration policy. We and build utilize 3D semantic maps to learn both action and perception in a completely self-supervised manner.
  arXiv  Detail & Related papers  (2021-12-02T06:26:38Z)
• Interesting Object, Curious Agent: Learning Task-Agnostic Exploration [44.18450799034677]
  In this paper, we propose a paradigm change in the formulation and evaluation of task-agnostic exploration. We show that our formulation is effective and provides the most consistent exploration across several training-testing environment pairs.
  arXiv  Detail & Related papers  (2021-11-25T15:17:32Z)
• Discovery-and-Selection: Towards Optimal Multiple Instance Learning for Weakly Supervised Object Detection [86.86602297364826]
  We propose a discoveryand-selection approach fused with multiple instance learning (DS-MIL) Our proposed DS-MIL approach can consistently improve the baselines, reporting state-of-the-art performance.
  arXiv  Detail & Related papers  (2021-10-18T07:06:57Z)
• MURAL: Meta-Learning Uncertainty-Aware Rewards for Outcome-Driven Reinforcement Learning [65.52675802289775]
  We show that an uncertainty aware classifier can solve challenging reinforcement learning problems. We propose a novel method for computing the normalized maximum likelihood (NML) distribution. We show that the resulting algorithm has a number of intriguing connections to both count-based exploration methods and prior algorithms for learning reward functions.
  arXiv  Detail & Related papers  (2021-07-15T08:19:57Z)
• Towards Open World Object Detection [68.79678648726416]
  ORE: Open World Object Detector is based on contrastive clustering and energy based unknown identification. We find that identifying and characterizing unknown instances helps to reduce confusion in an incremental object detection setting.
  arXiv  Detail & Related papers  (2021-03-03T18:58:18Z)
• Reannealing of Decaying Exploration Based On Heuristic Measure in Deep Q-Network [82.20059754270302]
  We propose an algorithm based on the idea of reannealing, that aims at encouraging exploration only when it is needed. We perform an illustrative case study showing that it has potential to both accelerate training and obtain a better policy.
  arXiv  Detail & Related papers  (2020-09-29T20:40:00Z)
• Co-training for On-board Deep Object Detection [0.0]
  Best performing deep vision-based object detectors are trained in a supervised manner by relying on human-labeled bounding boxes. Co-training is a semi-supervised learning method for self-labeling objects in unlabeled images. We show how co-training is a paradigm worth to pursue for alleviating object labeling, working both alone and together with task-agnostic domain adaptation.
  arXiv  Detail & Related papers  (2020-08-12T19:08:59Z)

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.