Edge Intelligence Workshop
March 2-3, 2020
SCIENTIFIC PROGRAM
The presentation files of the invited speakers and the pdf file of the presented posters are accessible from https://drive.google.com/drive/folders/1kaRXK0hTiKwBNN6JNi-7fcGisl0xhm7j?usp=sharing
| Monday 2 March 2020 | |||
|---|---|---|---|
| 08:30-09:00 | Yanhui Geng, Sébastien Le Digabel | Directors opening | |
| 09:00-09:30 | Mehdi Rezagholizadeh | Edge intelligence challenges in Huawei Noah's Ark Speech and Semantics Lab | |
| 09:30-10:00 | Bianca Schroeder | Machine learning and data mining for better computer systems | |
| 10:00-10:30 | Coffee Break | ||
| 10:30-11:00 | Alejandro Murua | Neural performance under compression | |
| 11:00-11:30 | Masoud Asgharian | On intrinsic dimension and causality | |
| 11:30-12:00 | Ashish Khisti | Information theoretic approaches in distributed machine learning | |
| 12:00-13:00 | Lunch box and Poster Session 1 | ||
| Nonvikan Karl-Augustt Alahassa; Alejandro Murua Shallow Structured Potts Neural Network Regression (S-SPNNR) Serge Vicente; Alejandro Murua Statistical learning with the determinantal point process Damoon Robatian; Masoud Asgharian Semi$^+$-supervised learning under sample selection bias Dipayan Mitra; Ashish Khisti Distributed Stochastic Gradient Decent with Quantized Compressive Sensing Alex Labach; Shahrokh Valaee Neural Network Sparsification Using Gibbs Measures Eyyüb Sari; Vahid Partovi Nia Batch Normalization in Quantized Networks |
Mohammad Javad Shafiee; Andrew Hryniowski; Francis Li; Zhong Qiu Lin; Alexander Wong State of Compact Architecture Search For Deep Neural Networks Adel Abusitta; Omar Abdul Wahab; Talal Halabi Deep learning for proactive cooperative malware detection System Vasileios Lioutas; Ahmad Rashid; Krtin Kumar On Compressing The Embedding Matrix Of Language Models For Edge Deployment Ramchalam Ramakrishnan; Eyyüb Sari; Vahid Partovi Nia Differentiable Mask for Pruning Convolutional and Recurrent Networks Bharat Venkitesh; Md. Akmal Haidar; Mehdi Rezagholizadeh On-Device Neural Text Segmentation for Augmented Reality Translation |
||
| 13:00-13:30 | Shahrokh Valaee | A unifying framework for dropout in neural networks | |
| 13:30-14:00 | Ali Ghodsi | Supervised Random Projections with Light | |
| 14:30-15:00 | Hassan Ashtiani | Learning through the Lens of Compression | |
| 15:00-15:30 | Coffee Break | ||
| 15:30-16:00 | Yingxue Zhang | Bayesian graph neural networks and its application in recommendation systems | |
| 16:00-16:30 | Nandita Vijaykumar | Hardware-software co-design for efficiency and programmability for sparse matrix operations | |
| 16:30-17:00 | Dominique Orban | Perspectives in computational optimization | |
| 17:00-17:30 | Charles Audet | Challenges and perspectives in blackbox optimization | |
| 17:30-18:00 | Sébastien Le Digabel | Blackbox optimization with the NOMAD solver | |
| Tuesday 3 March 2020 | |||
| 08:30-09:00 | Brett Meyer | Probabilistic sequential multi-objective optimization of convolutional neural networks | |
| 09:00-09:30 | Warren Gross | Stochastic computing for machine learning | |
| 09:30-10:00 | Eyal de Lara | System support for smart applications on the edge | |
| 10:00-10:30 | Coffee Break | ||
| 10:30-11:00 | Pascal Poupart | Diachronic embedding for temporal knowledge graph completion | |
| 11:00-11:30 | Tiago Falk | Signal processing for domain-enriched learning for speech applications | |
| 11:30-12:00 | Yaoliang Yu | Multi-objective federated learning: Convergence and robustness | |
| 12:00-13:00 | Lunch box and Poster Session 2 | ||
| Ji Xin; Raphael Tang; Jaejun Lee; Yaoliang Yu; Jimmy Lin Progress and Challenges in Early Exit for BERT Guojun Zhang; Yaoliang Yu Convergence of Gradient Methods on Bilinear Zero-Sum Games Kaiwen Wu; Allen Houze Wang; Yaoliang Yu Efficient Wasserstein Adversarial Attacks Ibtihel Amara; James Clark Uncertainty Transfer with Knowledge Distillation Ghouthi Boukli Hacene; Vincent Gripon; Matthieu Arzel; Nicolas Farrugia; Yoshua Bengio Pruning for Efficient Hardware Implementations of Deep Neural Networks |
Qing Tian; Tal Arbel; James Clark Deep LDA-Pruned Nets and their Robustness Zeou Hu; Yuxin Zhu; Ihab F Ilyas; Yaoliang Yu Fair Machine Learning through multi-objective optimization Xinlin Li; Vahid Partovi Nia Random Bias Initialization Improves Quantized Training Joao Felipe Santos; Tiago H Falk Pruning recurrent speech enhancement models via stuck neuron elimination Mahdi Zolnouri; Xinlin Li; Vahid Partovi Nia Importance of Data Loading Pipeline in Training Deep Neural Networks |
||
| 13:00-13:30 | Andrea Lodi | Neural networks and mixed-integer programming | |
| 13:30-14:00 | Daniel Aloise | A Lagrangean-based score for assessing the quality of pairwise constraints in semi-supervised clustering | |
| 14:30-15:00 | Mohan Liu | On-device: bringing artificial intelligence closer to consumer | |
| 15:00-15:30 | Coffee Break | ||
| 15:30-16:00 | Yoshua Bengio | Towards low-energy deep learning | |
| 16:00-16:30 | Vahid Partovi Nia | Edge intelligence challenges in Huawei Noah's Ark Computer Vision Lab | |
| 16:30-17:00 | James Clark | Task-dependent structured pruning of neural networks | |
| 17:00-17:30 | Fatiha Sadat | Natural language processing in low-resource settings | |
| 17:30-18:00 | Adam Oberman | From an ODE for Nesterov’s method to Accelerated SGD | |
| 18:00-18:30 | Murat Erdogdu | Convergence rates for diffusions-based sampling and optimization methods | |
| 18:30-19:00 | Organizers | Workshop wrap up | |
