Search Results for author:
Found 11 papers, 5 papers with code
Delta Keyword Transformer: Bringing Transformers to the Edge through Dynamically Pruned Multi-Head Self-Attention
Moreover, a reduction of ~87-94% operations can be achieved when only degrading the accuracy by 1-4%, speeding up the multi-head self-attention inference by a factor of ~7. 5-16.
ProbLP: A framework for low-precision probabilistic inference
Bayesian reasoning is a powerful mechanism for probabilistic inference in smart edge-devices.
Feed-Forward On-Edge Fine-tuning Using Static Synthetic Gradient Modules
By first training the SGMs in a meta-learning manner on a set of common objects, during fine-tuning, the SGMs provided the model with accurate gradients to successfully learn to grasp new objects.
ZigZag: A Memory-Centric Rapid DNN Accelerator Design Space Exploration Framework
This work introduces ZigZag, a memory-centric rapid DNN accelerator DSE framework which extends the DSE with uneven mapping opportunities, in which operands at shared memory levels are no longer bound to use the same memory levels for each loop index.
Distributed, Parallel, and Cluster Computing C.1.4; C.3; C.4
Benchmarking TinyML Systems: Challenges and Direction
In this position paper, we present the current landscape of TinyML and discuss the challenges and direction towards developing a fair and useful hardware benchmark for TinyML workloads.
Towards Hardware-Aware Tractable Learning of Probabilistic Models
We showcase our framework on a mobile activity recognition scenario, and on a variety of benchmark datasets representative of the field of tractable learning and of the applications of interest.
A multi-layered energy consumption model for smart wireless acoustic sensor networks
This model provides a first step of exploration prior to custom design of a smart wireless acoustic sensor, and also can be used to compare the energy consumption of different protocols.
BinarEye: An Always-On Energy-Accuracy-Scalable Binary CNN Processor With All Memory On Chip in 28nm CMOS
This paper introduces BinarEye: a digital processor for always-on Binary Convolutional Neural Networks.
Resource aware design of a deep convolutional-recurrent neural network for speech recognition through audio-visual sensor fusion
Performing multi-modal speech recognition - processing acoustic speech signals and lip-reading video simultaneously - significantly enhances the performance of such systems, especially in noisy environments.
Minimum Energy Quantized Neural Networks
To this end, the energy consumption of inference is modeled for a generic hardware platform.
Energy-Efficient ConvNets Through Approximate Computing
Recently ConvNets or convolutional neural networks (CNN) have come up as state-of-the-art classification and detection algorithms, achieving near-human performance in visual detection.
