Difference between revisions of "User:Fconti"
From iis-projects
Line 16: | Line 16: | ||
* [[Variable Bit Precision Logic for Deep Learning and Artificial Intelligence]] (master thesis) | * [[Variable Bit Precision Logic for Deep Learning and Artificial Intelligence]] (master thesis) | ||
* [[Embedded Audio Source Localization Exploiting Coincidence Detection in Asynchronous Spike Streams]] (semester/master thesis) | * [[Embedded Audio Source Localization Exploiting Coincidence Detection in Asynchronous Spike Streams]] (semester/master thesis) | ||
− | |||
* [[Towards Online Training of CNNs: Hebbian-Based Deep Learning]] (semester thesis) | * [[Towards Online Training of CNNs: Hebbian-Based Deep Learning]] (semester thesis) | ||
* [[Tiny CNNs for Ultra-Efficient Object Detection on PULP]] (semester thesis) | * [[Tiny CNNs for Ultra-Efficient Object Detection on PULP]] (semester thesis) | ||
<!-- * [[Single-Bit-Synapse Spiking Neural System-on-Chip]] (master thesis)--> | <!-- * [[Single-Bit-Synapse Spiking Neural System-on-Chip]] (master thesis)--> | ||
* You can also '''always''' contact me for projects that have not yet been formalized (semester/master), or with your own project idea! | * You can also '''always''' contact me for projects that have not yet been formalized (semester/master), or with your own project idea! | ||
+ | |||
+ | ==Projects in progress== | ||
+ | * [[Hardware Accelerators for Lossless Quantized Deep Neural Networks]] (semester/master thesis) | ||
==Past projects== | ==Past projects== |
Revision as of 21:12, 29 January 2019
Contents
Francesco Conti
I am a postdoctoral researcher here at IIS since July 2016, and I also hold a position as a research assistant at the University of Bologna, Italy. My interests focus on energy-efficient multicore architectures, with particular emphasis on heterogeneous multicores and cognitive computing architectures. I am interested in the entire technological stack "from Python to Silicon".
Interests
- heterogeneous multicore architectures for energy-efficient and low-power embedded systems
- applications of deep learning and brain-inspired computational models and architectures to low-power digital systems
Available projects
- Combining Pruning with Quantization for Best-Efficiency Neural Networks (semester/master thesis)
- Variable Bit Precision Logic for Deep Learning and Artificial Intelligence (master thesis)
- Embedded Audio Source Localization Exploiting Coincidence Detection in Asynchronous Spike Streams (semester/master thesis)
- Towards Online Training of CNNs: Hebbian-Based Deep Learning (semester thesis)
- Tiny CNNs for Ultra-Efficient Object Detection on PULP (semester thesis)
- You can also always contact me for projects that have not yet been formalized (semester/master), or with your own project idea!
Projects in progress
- Hardware Accelerators for Lossless Quantized Deep Neural Networks (semester/master thesis)
Past projects
- Low-Energy Cluster-Coupled Vector Coprocessor for Special-Purpose PULP Acceleration (semester/master thesis)
- Deep-Learning Based Phoneme Recognition from a Ultra-Low Power Spiking Cochlea (semester thesis)
- A Low-Power FPGA-Based Camera Sensor Interface for PULP Devices (semester thesis)
- A Recurrent Neural Network Speech Recognition Chip (semester thesis)
- Design and Implementation of Digital Spiking Neurons for Ultra-Low-Power In-Cluster Coprocessors (master thesis)
- Interfacing PULP with a Brain-Inspired Ultra-Low Power Spiking Cochlea (master thesis)
- Ultra-Efficient Visual Classification on Movidius Myriad2 (semester thesis)
Contact Information
- Office: ETZ J78
- telephone: +41 44 63 37151
- e-mail: fconti@iis.ee.ethz.ch
- www: Francesco Conti homepage @University of Bologna