Published in In progress, 2025
I am partnering with UW neuroscience researcher Ian Quah to investigate neuro science informed heuristics in the search for ultra energy efficient and performant edge vision transformers, through data dependent subnetwork computation.
Recommended citation: Forthcoming. Forthcoming
Published in In progress, 2025
Led by Cynthia Dong to support ranger safety and effectiveness through edge generative AI powered recording of field events from noisy speech to mobile form submission format.
Recommended citation: Forthcoming. Forthcoming
Published in In progress, 2025
Partnering with Kyle Johnson and Vicente Aroyos to enable battery free detection and classification of insects using solar powered milli robots and tiny ML for microcontrollers. We have a late breaking poster submission to ICRA with initial results.
Recommended citation: Forthcoming. Forthcoming
Published in In progress, 2025
We propose a learning algorithm for developing single tower on device edge mixture of experts models for hybrid vision transformers (convolution/transformer based), with downstream application towards efficient temporally-aligned multimodal sensor fusion in off-grid settings.
Recommended citation: Forthcoming. Forthcoming
Published in Under review IJCAI 2025, 2025
We propose a mixture of expert model which has subnetworks dedicated to detection animal species from specific geographical locations. This work builds towards a single species detector that prunes parts of the network not needed for a camera trap site without finetuning the model.
Recommended citation: Forthcoming. Forthcoming
Published in Under review ICML 2025, 2024
We explore per patch conditional computation for mobile vision transformers (vision only case), that will eventually be used for single-tower multimodal edge models.
Recommended citation: Mensah, Emmanuel Azuh, et al. "Towards Vision Mixture of Experts for Wildlife Monitoring on the Edge." arXiv preprint arXiv:2411.07834 (2024). https://arxiv.org/html/2411.07834v1
Published in Computing Within Limits, 2021
In the face of ecological and political limits, we propose a computational approach to opening borders to migrants in an automated way while ensuring safety for the host nation.
Recommended citation: Mensah, E. A., Singanamalla, S., Anderson, R., & Heimerl, K. (2021). When Borders Blur-Overcoming Political Limits with Computing in Truly Global Societies. https://computingwithinlimits.org/2021/papers/limits21-mensah.pdf
Published in Interspeech, 2019
This work presented a multimodal approach to learn bilingual lexicon directly from speech signals in two languages without the need for text by using vision as an interlingua. The approach starts a line of inquiry that can build word level translation between a pair of languages using say youtube videos that have similar objects but with speech in the two languages.
Recommended citation: Azuh, Emmanuel, David Harwath, and James R. Glass. "Towards Bilingual Lexicon Discovery From Visually Grounded Speech Audio." INTERSPEECH. 2019. http://groups.csail.mit.edu/sls/publications/2019/EmmanuelAzuh_Interspeech-2019.PDF
Published in IEEE Sensors, 2018
We use a deep autoencoder approach to learn representations of multivariate physiological signals that can be hashed and used to compute similarities between patients to assist in predicting critical events.
Recommended citation: Dhamala, J., Azuh, E., Al-Dujaili, A., Rubin, J., & O’Reilly, U. M. (2018). Multivariate time-series similarity assessment via unsupervised representation learning and stratified locality sensitive hashing: Application to early acute hypotensive episode detection. IEEE Sensors Letters, 3(1), 1-4. https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=8506445