Papers › Generative Pretrained Embedding and Hierarchical Irregular Time Series Representation...

Generative Pretrained Embedding and Hierarchical Irregular Time Series Representation for Daily Living Activity Recognition

27 Dec 2024arXiv:2412.19732archive 2025-07-28

Damien Bouchabou, Sao Mai Nguyen

Within the evolving landscape of smart homes, the precise recognition of daily living activities using ambient sensor data stands paramount. This paper not only aims to bolster existing algorithms by evaluating two distinct pretrained embeddings suited for ambient sensor activations but also introduces a novel hierarchical architecture. We delve into an architecture anchored on Transformer Decoder-based pre-trained embeddings, reminiscent of the GPT design, and contrast it with the previously established state-of-the-art (SOTA) ELMo embeddings for ambient sensors. Our proposed hierarchical structure leverages the strengths of each pre-trained embedding, enabling the discernment of activity dependencies and sequence order, thereby enhancing classification precision. To further refine recognition, we incorporate into our proposed architecture an hour-of-the-day embedding. Empirical evaluations underscore the preeminence of the Transformer Decoder embedding in classification endeavors. Additionally, our innovative hierarchical design significantly bolsters the efficacy of both pre-trained embeddings, notably in capturing inter-activity nuances. The integration of temporal aspects subtly but distinctively augments classification, especially for time-sensitive activities. In conclusion, our GPT-inspired hierarchical approach, infused with temporal insights, outshines the SOTA ELMo benchmark.

PaperPDFCode

Code

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Activity RecognitionDecoderIrregular Time SeriesTime Series

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

Absolute Position EncodingsAdamAttentionAttention DropoutBPEBiLSTMCosine AnnealingDense ConnectionsDiscriminative Fine-TuningDropoutELMoGPTLSTMLabel SmoothingLayer NormalizationLinear LayerLinear Warmup With Cosine AnnealingMulti-Head AttentionPosition-Wise Feed-Forward LayerResidual ConnectionSigmoid ActivationSoftmaxTanh ActivationTransformerWeight Decay

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections