Recurrent Neural Network for MoonBoard Climbing Route Classification and Generation

• URL: http://arxiv.org/abs/2102.01788v1
• Date: Tue, 2 Feb 2021 22:38:23 GMT
• Title: Recurrent Neural Network for MoonBoard Climbing Route Classification and Generation
• Authors: Yi-Shiou Duh, Ray Chang
• Abstract summary: "BetaMove" is a new move preprocessing pipeline developed in order to mimic a human climber's hand sequence. The accuracy of our grade predictor reaches near human-level performance. Our route generator produces new routes of much better quality compared to previous work.
• Score: 0.0
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• Abstract: Classifying the difficulties of climbing routes and generating new routes are both challenging. Existing machine learning models not only fail to accurately predict a problem's difficulty, but they are also unable to generate reasonable problems. In this work, we introduced "BetaMove", a new move preprocessing pipeline we developed, in order to mimic a human climber's hand sequence. The preprocessed move sequences were then used to train both a route generator and a grade predictor. By preprocessing a MoonBoard problem into a proper move sequence, the accuracy of our grade predictor reaches near human-level performance, and our route generator produces new routes of much better quality compared to previous work. We demonstrated that with BetaMove, we are able to inject human insights into the machine learning problems, and this can be the foundations for future transfer learning on climbing style classification problems.

Related papers

• Using Machine Learning for move sequence visualization and generation in climbing [35.1762496625647]
  We develop a visualization tool for move sequence evaluation on a given boulder. Then, we look into move sequence prediction from simple holds sequence information using three different Transformer models.
  arXiv  Detail & Related papers  (2025-03-01T11:50:36Z)
• One-Layer Transformer Provably Learns One-Nearest Neighbor In Context [48.4979348643494]
  We study the capability of one-layer transformers learning the one-nearest neighbor rule. A single softmax attention layer can successfully learn to behave like a one-nearest neighbor.
  arXiv  Detail & Related papers  (2024-11-16T16:12:42Z)
• Learning Humanoid Locomotion over Challenging Terrain [84.35038297708485]
  We present a learning-based approach for blind humanoid locomotion capable of traversing challenging natural and man-made terrains. Our model is first pre-trained on a dataset of flat-ground trajectories with sequence modeling, and then fine-tuned on uneven terrain using reinforcement learning. We evaluate our model on a real humanoid robot across a variety of terrains, including rough, deformable, and sloped surfaces.
  arXiv  Detail & Related papers  (2024-10-04T17:57:09Z)
• Humanoid Locomotion as Next Token Prediction [84.21335675130021]
  Our model is a causal transformer trained via autoregressive prediction of sensorimotor trajectories. We show that our model enables a full-sized humanoid to walk in San Francisco zero-shot. Our model can transfer to the real world even when trained on only 27 hours of walking data, and can generalize commands not seen during training like walking backward.
  arXiv  Detail & Related papers  (2024-02-29T18:57:37Z)
• Board-to-Board: Evaluating Moonboard Grade Prediction Generalization [0.0]
  Bouldering is a sport where athletes aim to climb up an obstacle using a set of defined holds called a route. The variation in individual climbers technical and physical attributes and many nuances of an individual route make grading a difficult and often biased task. We apply classical and deep-learning modelling techniques to the 2016, 2017 and 2019 Moonboard datasets.
  arXiv  Detail & Related papers  (2023-11-21T08:16:01Z)
• Towards Machine Learning for Placement and Routing in Chip Design: a Methodological Overview [72.79089075263985]
  Placement and routing are two indispensable and challenging (NP-hard) tasks in modern chip design flows. Machine learning has shown promising prospects by its data-driven nature, which can be of less reliance on knowledge and priors.
  arXiv  Detail & Related papers  (2022-02-28T06:28:44Z)
• Learning to Shift Attention for Motion Generation [55.61994201686024]
  One challenge of motion generation using robot learning from demonstration techniques is that human demonstrations follow a distribution with multiple modes for one task query. Previous approaches fail to capture all modes or tend to average modes of the demonstrations and thus generate invalid trajectories. We propose a motion generation model with extrapolation ability to overcome this problem.
  arXiv  Detail & Related papers  (2021-02-24T09:07:52Z)
• Learning Stable Manoeuvres in Quadruped Robots from Expert Demonstrations [3.893720742556156]
  Key problem is to generate leg trajectories for continuously varying target linear and angular velocities. We propose a two pronged approach to address this problem. We develop a neural network-based filter that takes in target velocity, radius and transforms them into new commands.
  arXiv  Detail & Related papers  (2020-07-28T15:02:04Z)
• Human Motion Transfer from Poses in the Wild [61.6016458288803]
  We tackle the problem of human motion transfer, where we synthesize novel motion video for a target person that imitates the movement from a reference video. It is a video-to-video translation task in which the estimated poses are used to bridge two domains. We introduce a novel pose-to-video translation framework for generating high-quality videos that are temporally coherent even for in-the-wild pose sequences unseen during training.
  arXiv  Detail & Related papers  (2020-04-07T05:59:53Z)

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.