WeAreDevelopers Logo WeAreDevelopers
Sign in
For You LIVE
WeAreDevelopers LIVE • Mar 23, 2022

Intelligent Data Selection for Continual Learning of AI Functions

Nico Schmidt

Stop wasting bandwidth on random fleet data. Learn how intelligent edge filtering isolates low-confidence edge cases, drastically reducing overhead while accelerating autonomous AI.

Pause
Mute Enter Fullscreen
#1 about 4 min

Introduction to active learning and continuous data selection

The core use cases and definitions behind actively selecting diverse datasets for machine learning functions.

#2 about 4 min

Evaluating edge data sources and automotive compute capabilities

The constraints and benefits of using static data lakes versus real-time testing and customer vehicle fleets.

#3 about 3 min

Capturing informative data targets from the long tail

How targeting rare traffic scenarios, unusual sensor noise, and underrepresented classes improves model accuracy.

#4 about 4 min

Methodologies for mapping prediction uncertainty and structural anomalies

Techniques for estimating prediction certainty and detecting new operational density patterns.

#5 about 2 min

Utilizing softmax uncertainty for vehicular traffic light detection

Using native score aggregations to capture bounding box errors and weak signal situations.

#6 about 5 min

Benchmarking active learning algorithms against random data sampling

Why evaluating selective triggering demands customized corner-case test datasets beyond standard baseline measurements.

#7 about 1 min

Developing verifiable safety metrics for automated driving perception

Defining standardized performance metrics to assure the reliable safety operations of automotive perception algorithms.

#8 about 4 min

Deploying intelligent edge filters to onboard vehicular modules

Using the in-car instinct module to filter high-value detection snippets for cloud upload before refining production models.

#9 about 3 min

Adapting vehicular perception models to new geographical domains

Applying data learning concepts and targeted upload triggers to bridge environmental gaps between regional deployments.

#10 about 3 min

Architecting a universal plugin framework for data selection

Constructing an independent module topology that accepts logical, diversity, and novelty-based algorithms.

#11 about 2 min

Exporting framework independent representations for edge processing platforms

Exporting distinct training models to neutral ONNX formats and accelerating inference pipelines to run across disparate edge processing units.

#12 about 2 min

Bridging Python prototyping environments and embedded systems languages

Overcoming prototyping barriers by optimizing zero-copy and inter-process communications for efficient onboard execution.

#13 about 4 min

Navigating software integration compliance in safety automotive environments

Implementing open source license validations, strict code quality benchmarks, and complete requirements traceability.

#14 about 3 min

Decoupling cloud workflows from specific machine learning frameworks

Emphasizing agnostic pipelines that support mixed stacks without enforcing rigid cloud dependencies.

#15 about 3 min

Gauging specific model improvement metrics through custom situations

Examining performance distributions relative to data selection approaches and evaluating dataset composition bounds.

#16 about 5 min

Monitoring production regression using side loaded analysis logic

Establishing metrics alongside edge production systems to verify that machine learning inputs do not drift.

#17 about 4 min

Balancing data science skillings alongside systems engineering rigor

Evaluating candidate profiles for rigorous systems engineering aptitude within active data operations.

#18 about 3 min

Incorporating automated machine learning optimizations into hardware loops

Weighing the cost benefits of architectural searches against the constraints of embedded automotive microcontrollers.

#19 about 2 min

Distinguishing extreme system outliers from boundary line predictions

Why isolating unexpected system events requires merging diversity scoring frameworks with raw uncertainty logic.

Matching moments

2:52 min

Introduction to safety-critical machine learning in automotive contexts

Jan Zawadzki · WWC 2022

4:55 min

Audience Q&A on autonomous driving models and data

Liang Yu · WWC 2022

2:25 min

Navigating automotive complexity with AI runtime environments

Daniel Graff +1 · WWC 2021

3:37 min

Transitioning automated driving to neural networks and model-based perception

Katrin Lehmann Katrin Lehmann +1 · WWC 2025

1:29 min

The virtuous cycle of machine learning in connected cars

Jan Zawadzki · LIVE

1:43 min

Optimizing AI model execution for in-car inference

Daniel Graff +1 · WWC 2021

More from WeAreDevelopers LIVE

Aliens in Quarantine, Python Deprecates Loops, and AI Buys Old Books - Zbyszek Tenerowicz thumbnail
5:43
5 days ago

Aliens in Quarantine, Python Deprecates Loops, and AI Buys Old Books - Zbyszek Tenerowicz

Chris Heilmann, Daniel Cranney

WeAreDevelopers LIVE - Node and Package Security thumbnail
1:08:30
5 days ago

WeAreDevelopers LIVE - Node and Package Security

Chris Heilmann, Daniel Cranney, Zbyszek Tenerowicz

WeAreDevelopers LIVE - Streaming HTML thumbnail
1:10:26
12 days ago

WeAreDevelopers LIVE - Streaming HTML

Chris Heilmann, Daniel Cranney, Jad Joubran

WeAreDevelopers LIVE - CODE100 at WeAreDevelopers Europe thumbnail
1:02:34
19 days ago

WeAreDevelopers LIVE - CODE100 at WeAreDevelopers Europe

Chris Heilmann, Daniel Cranney, Justin Halsall, Werner Jeschko, Nicole Jeschko, Mirco Tomić

WeAreDevelopers LIVE - Web Performance in an AI World thumbnail
1:07:59
26 days ago

WeAreDevelopers LIVE - Web Performance in an AI World

Chris Heilmann, Daniel Cranney

FIFA Hack, Midjourney Does Ultrasounds, and Trampoline Traffic Lights - Nicole Jeschko and Werner Jeschko thumbnail
5:55
about 2 months ago

FIFA Hack, Midjourney Does Ultrasounds, and Trampoline Traffic Lights - Nicole Jeschko and Werner Jeschko

Chris Heilmann, Daniel Cranney

Upcoming sessions on this topic

Open session

World Congress 2026 North America

You Can’t Re-Run Sunlight: Designing ML Data Architectures for Physical AI

An Phan

Senior Data Infrastructure Engineer @ Hippo Harvest

An Phan
Open session

World Congress 2026 North America

Closing the Visibility Gap: Lessons from Safety Critical Agentic Systems

Vivek Pandit

Principal Engineer at Cadence

Vivek Pandit
Open session

World Congress 2026 North America

Trust, But Verify: Continuous GPU Validation at Scale

Kyle Bell

VP of AI @ TensorWave

Kyle Bell
Open session

World Congress 2026 North America

Agents That Own Their Inference: Building Production AI Agents on Dedicated GPUs

Duan Lightfoot

Sr. AI Engineer, Akamai

Duan Lightfoot
Open session

World Congress 2026 North America

Proactive AI That Doesn’t Annoy Users: Building Context-Aware Notification Systems

Raju Dandigam Dandigam

Engineering Manager at Navan

Raju Dandigam Dandigam
Open session

World Congress 2026 North America

No Single Model to Rule Them All: Building Resilient AI Agents Across Open & Closed LLMs

Emmanuel Acheampong

Senior Manager Developer Relations at Crusoe AI

Emmanuel Acheampong

Related videos

What non-automotive Machine Learning projects can learn from automotive Machine Learning projects thumbnail
48:01
WWC 2022 • about 4 years ago

What non-automotive Machine Learning projects can learn from automotive Machine Learning projects

Jan Zawadzki

Finding the unknown unknowns: intelligent data collection for autonomous driving development thumbnail
19:24
WWC 2022 • about 4 years ago

Finding the unknown unknowns: intelligent data collection for autonomous driving development

Liang Yu

Uncertainty Estimation of Neural Networks thumbnail
1:00:28
WWC 2021 • about 5 years ago

Uncertainty Estimation of Neural Networks

Tillman Radmer, Fabian Hüger, Nico Schmidt

How Machine Learning is turning the Automotive Industry upside down thumbnail
26:55
LIVE • almost 6 years ago

How Machine Learning is turning the Automotive Industry upside down

Jan Zawadzki

Developing an AI.SDK thumbnail
38:12
WWC 2021 • about 5 years ago

Developing an AI.SDK

Daniel Graff, Andreas Wittmann

Getting Started with Machine Learning thumbnail
44:11
LIVE • almost 5 years ago

Getting Started with Machine Learning

Alexandra Waldherr

Related articles

From learning to earning

Jobs that call for the skills explored in this talk.