Bayesian Optimization of Neural Architectures for Human Activity Recognition

Osmani, Aomar; Hamidi, Massinissa

doi:10.1007/978-3-030-13001-5_12

Cited by 5 publications

(5 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To compare our baseline with the proposed hierarchical model, we make sure to get the same complexity, i.e., comparable number of parameters as the largest hierarchies including the weights of the learners. We also use Bayesian optimization based on Gaussian processes as surrogate models to select the optimal hyperparameters of the baseline model (Snoek, Larochelle, and Adams 2012;Osmani and Hamidi 2019). In addition, we compare our proposed approach with the following closely related baselines from the HAR literature:…”

Section: Methodsmentioning

confidence: 99%

Clustering Approach to Solve Hierarchical Classification Problem Complexity

Osmani¹,

Hamidi²,

Alizadeh

2022

AAAI

Self Cite

View full text Add to dashboard Cite

In a large domain of classification problems for real applications, like human activity recognition, separable spaces between groups of concepts are easier to learn than each concept alone. This is because the search space biases required to separate groups of classes (or concepts) are more relevant than the ones needed to separate classes individually. For example, it is easier to learn the activities related to the body movements group (running, walking) versus "on-wheels" activities group (bicycling, driving a car), before learning more specific classes inside each of these groups. Despite the obvious interest of this approach, our theoretical analysis shows a high complexity for finding an exact solution. We propose in this paper an original approach based on the association of clustering and classification approaches to overcome this limitation. We propose a better approach to learn the concepts by grouping classes recursively rather than learning them class by class. We introduce an effective greedy algorithm and two theoretical measures, namely cohesion and dispersion, to evaluate the connection between the clusters and the classes. Extensive experiments on the SHL dataset show that our approach improves classification performances while reducing the number of instances used to learn each concept.

show abstract

Section: Methodsmentioning

confidence: 99%

Clustering Approach to Solve Hierarchical Classification Problem Complexity

Osmani¹,

Hamidi²,

Alizadeh

2022

AAAI

Self Cite

View full text Add to dashboard Cite

show abstract

“…For instance, in Nguyen, Liu, and Zio (2020) an accurate and reliable multi-step ahead prediction model based on LSTM, whose hyperparameters have been optimized through BO, has been validated on steam generator data acquired from different French nuclear power plants for prognostic and health management of the plants. In Osmani and Hamidi (2019), an expensive and time-consuming design of a Deep Neural Network for human activity recognition has been addressed via BO in order to optimally and efficiently tune the deep neural architectures' hyper-parameters. With respect to ADS, a recent and interesting application of BO is devoted to generate simulation scenarios in order to improve accuracy and "safety" of the ADS (Abeysirigoonawardena, Shkurti, and Dudek, 2019;Zerwas, Kalmbach, Henkel, Rétvári, Kellerer, Blenk, and Schmid, 2019;Gangopadhyay, Khastgir, Dey, Dasgupta, Montana, and Jennings, 2019).…”

Section: Related Workmentioning

confidence: 99%

Bayesian Optimization and Deep Learning forsteering wheel angle prediction

Riboni¹,

Ghioldi²,

Candelieri³

et al. 2021

Preprint

View full text Add to dashboard Cite

Automated driving systems (ADS) have undergone a significant improvement in the last years. ADS and more precisely self-driving cars technologies will change the way we perceive and know the word of transportation systems in terms of user experience, mode choices and business models. The emerging field of Deep Learning (DL) has been successfully applied for the development of innovative ADS solutions. However, the attempt to single out the best deep neural network architecture and tuning its hyperparameters are all expensive processes, both in terms of time and computational resources. In this work, Bayesian Optimization (BO) is used to optimize the hyperparameters of a Spatiotemporal-Long Short Term Memory (ST-LSTM) network with the aim to obtain an accurate model for the prediction of the steering angle in a ADS. BO was able to identify, within a limited number of trials, a model -namely BO ST-LSTM -which resulted, on a public dataset, the most accurate when compared to classical end-to-end driving models.

show abstract

“…Hyperparameters of the architectural components and their associated ranges used for the construction of the surrogate model. From[109].…”

mentioning

confidence: 99%

“…Architectures with recurrent output layers are referred to as Hybrid while those with dense output layers are referred to as Convolutional. From[109].…”

mentioning

confidence: 99%

Human Activity Recognition: A Dynamic Inductive Bias Selection Perspective

Hamidi¹,

Osmani²

2021

Sensors

Self Cite

View full text Add to dashboard Cite

In this article, we study activity recognition in the context of sensor-rich environments. In these environments, many different constraints arise at various levels during the data generation process, such as the intrinsic characteristics of the sensing devices, their energy and computational constraints, and their collective (collaborative) dimension. These constraints have a fundamental impact on the final activity recognition models as the quality of the data, its availability, and its reliability, among other things, are not ensured during model deployment in real-world configurations. Current approaches for activity recognition rely on the activity recognition chain which defines several steps that the sensed data undergo: This is an inductive process that involves exploring a hypothesis space to find a theory able to explain the observations. For activity recognition to be effective and robust, this inductive process must consider the constraints at all levels and model them explicitly. Whether it is a bias related to sensor measurement, transmission protocol, sensor deployment topology, heterogeneity, dynamicity, or stochastic effects, it is essential to understand their substantial impact on the quality of the data and ultimately on activity recognition models. This study highlights the need to exhibit the different types of biases arising in real situations so that machine learning models, e.g., can adapt to the dynamicity of these environments, resist sensor failures, and follow the evolution of the sensors’ topology. We propose a metamodeling approach in which these biases are specified as hyperparameters that can control the structure of the activity recognition models. Via these hyperparameters, it becomes easier to optimize the inductive processes, reason about them, and incorporate additional knowledge. It also provides a principled strategy to adapt the models to the evolutions of the environment. We illustrate our approach on the SHL dataset, which features motion sensor data for a set of human activities collected in real conditions. The obtained results make a case for the proposed metamodeling approach; noticeably, the robustness gains achieved when the deployed models are confronted with the evolution of the initial sensing configurations. The trade-offs exhibited and the broader implications of the proposed approach are discussed with alternative techniques to encode and incorporate knowledge into activity recognition models.

show abstract

Bayesian Optimization of Neural Architectures for Human Activity Recognition

Cited by 5 publications

References 15 publications

Clustering Approach to Solve Hierarchical Classification Problem Complexity

Clustering Approach to Solve Hierarchical Classification Problem Complexity

Bayesian Optimization and Deep Learning forsteering wheel angle prediction

Human Activity Recognition: A Dynamic Inductive Bias Selection Perspective

Contact Info

Product

Resources

About