Vittorio Mazzia (0000-0002-7624-1850)

Personal information

Biography

I'm a research scientist at Amazon AGI. Previously, at Politecnico di Torino as a Ph.D. student, working in synergy with the two interdepartmental centers, PIC4SeR and SmartData.

I spent my Ph.D. years researching novel deep learning algorithms for enhanced artificial perception, autonomous navigation, path planning, and sensorimotor solutions for service robotics. Furthermore, I worked on the efficiency of machine learning algorithms for embedded AI at the edge, using neural compute devices for hardware acceleration.

My current research focuses on delving into the capacities and limitations of large neural models. Specifically, I am intrigued by the intricate process through which sensory signals contribute to shaping a comprehensive model of the world, and the potential role this model can play in mitigating existing limitations within large-scale models. Always, neural efficiency and generalization are my driving principles.

Activities

Collapse all

Amazon: WA, WA, US

2024-01-01 to present | Research Scientist II (AGI)

Employment

Vittorio Mazzia

https://orcid.org/0000-0002-7624-1850

Personal information

Websites & social links

Other IDs

Keywords

Countries

Biography

Activities

Employment (3)

Amazon: WA, WA, US

Politecnico di Torino: Torino, Piemonte, IT

Amazon (United States): Turin, Piedmont, IT

Education and qualifications (2)

Politecnico di Torino Dipartimento di Elettronica e Telecomunicazioni: Torino, Piemonte, IT

Politecnico di Torino Dipartimento di Automatica e Informatica: Torino, Piemonte, IT

Works (23)

A Deep Learning Driven Algorithmic Pipeline for Autonomous Navigation in Row-Based Crops

Towards Comprehensive Subgroup Performance Analysis in Speech Models

Back-to-Bones: Rediscovering the role of backbones in domain generalization

A Survey on Knowledge Editing of Neural Networks

Exploring Subgroup Performance in End-to-End Speech Models

Action Transformer: A self-attention model for short-time pose-based human action recognition

Indoor Point-to-Point Navigation with Deep Reinforcement Learning and Ultra-Wideband

Efficient-CapsNet: capsule network with self-attention routing

Deep Semantic Segmentation at the Edge for Autonomous Navigation in Vineyard Rows

An Adaptive Row Crops Path Generator with Deep Learning Synergy

Robust ultra-wideband range error mitigation with deep learning at the edge

Domain-Adversarial Training of Self-Attention-Based Networks for Land Cover Classification Using Multi-Temporal Sentinel-2 Satellite Imagery

DeepWay: A Deep Learning waypoint estimator for global path generation

Autonomous Navigation in Vineyards with Deep Learning at the Edge

Real-Time Apple Detection System using Embedded Systems with Hardware Accelerators: an Edge AI Application

A Cost-Effective Person-Following System for Assistive Unmanned Vehicles with Deep Learning at the Edge

Multi-Image Super Resolution of Remotely Sensed Images Using Residual Attention Deep Neural Networks

Use of Deep Learning for Automatic Detection of Cracks in Tunnels

Local Motion Planner for Autonomous Navigation in Vineyards with a RGB-D Camera-Based Algorithm and Deep Learning Synergy

UAV and Machine Learning Based Refinement of a Satellite-Driven Vegetation Index for Precision Agriculture

Deep Learning Algorithms for Complex Pattern Recognition in Ultrasonic Sensors Arrays

Improvement in Land Cover and Crop Classification based on Temporal Features Learning from Sentinel-2 Data Using Recurrent-Convolutional Neural Network (R-CNN)

Refining satellite imagery by using UAV imagery for vineyard environment: A CNN Based approach