( events)   Timezone: »  
Mon Dec 13 07:50 AM -- 06:30 PM (PST)
Machine Learning for Autonomous Driving
Xinshuo Weng · Jiachen Li · Nick Rhinehart · Daniel Omeiza · Ali Baheri · Rowan McAllister

We propose a full-day workshop, called “Machine Learning for Autonomous Driving” (ML4AD), as a venue for machine learning (ML) researchers to discuss research problems concerning autonomous driving (AD). Our goal is to promote ML research, and its real-world impact, on self-driving technologies. Full self-driving capability (“Level 5”) is far from solved and extremely complex, beyond the capability of any one institution or company, necessitating larger-scale communication and collaboration, which we believe workshop formats help provide.

We propose a large-attendance talk format of approximately 500 attendees, including (1) a call for papers with poster sessions and spotlight presentations; (2) keynote talks to communicate the state-of-the-art; (3) panel debates to discuss future research directions; (4) a call for challenge to encourage interaction around a common benchmark task; (5) social breaks for newer researchers to network and meet others.

Opening Remarks
Reinforcement Learning for Autonomous Driving (Keynote Talk)
Q&A: Jeff Schneider (Live Q/A)
AV2.0: Deploying End to End Deep Learning Policies at Fleet Scale (Keynote Talk)
Q&A: Alex Kendall (Live Q/A)
(Best Paper) UMBRELLA: Uncertainty-Aware Model-Based Offline Reinforcement Learning Leveraging Planning (Oral)
Poster Session and Social
Physics-Guided AI for Modeling Autonomous Vehicle Dynamics (Keynote Talk)
Q&A: Rose Yu (Live Q/A)
The Ongoing Research in University of Michigan & Ford Center for Autonomous Vehicles (FCAV) (Keynote Talk)
Q&A: Matthew Johnson-Roberson (Live Q/A)
CARLA Challenge (Challenge)
Fantastic Failures and Where to Find Them: Designing Safe, Robust Autonomy (Keynote Talk)
Q&A: Katie Driggs-Campbell (Live Q/A)
Safely Learning Behaviors of Other Agents (Keynote Talk)
Q&A: Claire Tomlin (Live Q/A)
Spotlight Talks (Oral)
Poster Session and Social
Learning Driving Agents from Simulation (Keynote Talk)
Q&A: Mark Palatucci (Live Q/A)
Autonomous Vehicle Decision-Making Policy Fast Adaptation Using Meta Reinforcement Learning (Keynote Talk)
Q&A: Songan Zhang (Live Q/A)
Robotics for an ML-Driven World (Keynote Talk)
Q&A: Sarah Tang (Live Q/A)
Shifts: A Dataset of Real Distributional Shift Across Multiple Large-Scale Tasks (Challenge)
Closing Remarks
Object-Level Targeted Selection via Deep Template Matching (Poster)
MTL-TransMODS: Cascaded Multi-Task Learning for Moving Object Detection and Segmentation with Unified Transformers (Poster)
Offline Reinforcement Learning for Autonomous Driving with Safety and Exploration Enhancement (Poster)
Self-supervised Sun Glare Detection CNN for Self-aware Autonomous Driving (Poster)
Spatial-Temporal Gated Transformersfor Efficient Video Processing (Poster)
Efficient Unknown Object Detection with Discrepancy Networks for Semantic Segmentation (Poster)
Watch out for the risky actors: Assessing risk in dynamic environments for safe driving (Poster)
TITRATED: Learned Human Driving Behavior without Infractions via Amortized Inference (Poster)
UMBRELLA: Uncertainty-Aware Model-Based Offline Reinforcement Learning Leveraging Planning (Poster)
NSS-VAEs: Generative Scene Decomposition for Visual Navigable Space Construction (Poster)
Meta Guided Metric Learner for Overcoming Class Confusion in Few-Shot Road Object Detection (Poster)
Circular-Symmetric Correlation Layer (Poster)
Real-time Generalized Sensor Fusion with Transformers (Poster)
Switching Recurrent Kalman Networks (Poster)
Improved Object Detection in Thermal Imaging Through Context Enhancement and Information Fusion: A Case Study in Autonomous Driving (Poster)
Fast Polar Attentive 3D Object Detection based on Point Cloud (Poster)
PolyTrack: Tracking with Bounding Polygons (Poster)
How Far Can I Go ? : A Self-Supervised Approach for Deterministic Video Depth Forecasting (Poster)
Are Socially-Aware Trajectory Prediction Models Really Socially-Aware? (Poster)
Compressing Sensor Data for Remote Assistance of Autonomous Vehicles using Deep Generative Models (Poster)
ORDER: Open World Object Detection on Road Scenes (Poster)
Temporal Transductive Inference for Few-Shot Video Object Segmentation (Poster)
PKCAM: Previous Knowledge Channel Attention Module (Poster)
Scalable Primitives for Generalized Sensor Fusion in Autonomous Vehicles (Poster)
Does Thermal data make the detection systems more reliable? (Poster)
A Scenario-Based Platform for Testing Autonomous Vehicle Behavior Prediction Models in Simulation (Poster)
DriverGym: Democratising Reinforcement Learning for Autonomous Driving (Poster)
Reinforcement Learning as an Alternative to Reachability Analysis for Falsification of AD Functions (Poster)
AA3DNet: Attention Augmented Real Time 3D Object Detection (Poster)
Monocular 3D Object Detection by Leveraging Self-Supervised Visual Pre-training (Poster)
Self-Supervised Pretraining for Scene Change Detection (Poster)
Hierarchical Adaptable and Transferable Networks (HATN) for Driving Behavior Prediction (Poster)
A Step Towards Efficient Evaluation of Complex Perception Tasks in Simulation (Poster)
Incorporating Voice Instructions in Model-Based Reinforcement Learning for Self-Driving Cars (Poster)