Abstract
Understanding of human intention by observing a series of human actions has been a challenging task. In order to do so, we need to analyze longer sequences of human actions related with intentions and extract the context from the dynamic features. The multiple timescales recurrent neural network (MTRNN) model, which is believed to be a kind of solution, is a useful tool for recording and regenerating a continuous signal for dynamic tasks. However, the conventional MTRNN suffers from the vanishing gradient problem which renders it impossible to be used for longer sequence understanding. To address this problem, we propose a new model named Continuous Timescale Long-Short Term Memory (CTLSTM) in which we inherit the multiple timescales concept into the Long-Short Term Memory (LSTM) recurrent neural network (RNN) that addresses the vanishing gradient problem. We design an additional recurrent connection in the LSTM cell outputs to produce a time-delay in order to capture the slow context. Our experiments show that the proposed model exhibits better context modeling ability and captures the dynamic features on multiple large dataset classification tasks. The results illustrate that the multiple timescales concept enhances the ability of our model to handle longer sequences related with human intentions and hence proving to be more suitable for complex tasks, such as intention recognition.
Highlights
In machine learning, dynamic sequence modeling is a burning research topic, which includes intention understanding, action recognition, language understanding, semantic understanding (Peniak et al, 2011; Wasser and Lincoln, 2012; Wonmin et al, 2015; Kim et al, 2017) etc
We propose a model considering the advantages of an Long-Short Term Memory (LSTM) and inheriting the biological idea given by Continuous Recurrent Neural Network (CTRNN)
Inspired by multiple timescales recurrent neural network (MTRNN) and LSTM, we aim to develop a recurrent neural network (RNN) with multiple timescales structure with better ability to capture the dynamic features in longer sequences such as a series of human actions for understanding human intention
Summary
Dynamic sequence modeling is a burning research topic, which includes intention understanding, action recognition, language understanding, semantic understanding (Peniak et al, 2011; Wasser and Lincoln, 2012; Wonmin et al, 2015; Kim et al, 2017) etc. Context, which is generally mentioned in language understanding (Ghadessy, 1999; Givón, 2005), plays an important role in dynamic sequence classification. Context contains several physical and abstract aspects such as time, symbols, location, names, etc. Same words may have different meaning under different contexts. Context plays the role of surroundings, which contains some inconspicuous but important descriptions of the current phenomenon. Context can be deemed as the key of the dynamic sequence learning
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.