Safely Managing Generative AI with Humans in the Loop
Explore design patterns for human-in-the-loop approaches to ensure safe and trustworthy generative AI systems. Examine real-world case studies and learn valuable lessons for responsible AI implementation.
Understanding Where Generative AI Fits into Business
Explore the business applications and potential risks of Generative AI with insights from Deloitte experts on implementation and impact.
Towards Trustworthy Automatic Diagnosis Systems by Emulating Doctors' Reasoning with Deep Reinforcement Learning
Explore deep reinforcement learning for automated medical diagnosis, emulating doctors' reasoning to generate differential diagnoses, prioritize severe pathologies, and enhance system trustworthiness.
The Fairness of Fairness: Evaluating, Monitoring, and Impacts of ML in Insurance
Explore strategies for creating fair ML models in insurance, covering bias metrics, optimization techniques, and insights from FTC research on credit-based insurance scores' impact.
Composable Interventions for Language Models
Explore composable interventions for language models, enhancing accuracy and mitigating harmful outputs. Learn about interactions between different intervention types and their practical implications.
How to Steal ChatGPT's Embedding Size and Other Low-rank Logit Tricks
Uncover hidden information about API-protected LLMs through low-rank logit analysis. Learn to estimate embedding sizes, audit model updates, and identify source models efficiently.
Informative Example Selection for In-Context Learning
Explore efficient approaches for selecting informative examples in In-Context Learning, enhancing large language model performance without additional training.
Stop Explaining Black Box ML Models for High Stakes Decisions and Use Interpretable Models
Explore the societal impact of black box ML models in high-stakes decisions and discover the benefits of using interpretable models for transparent decision-making.
Minimize Risk and Accelerate MLOps with ML Monitoring and Explainability
Discover strategies to minimize risk and accelerate MLOps through ML monitoring and explainability. Learn to efficiently monitor, troubleshoot, and document ML models for improved performance and compliance.
Implementing a Machine Learning Initiative: A Leader's Perspective
Discover key strategies for successfully implementing machine learning initiatives in organizations. Learn to align ML with strategic objectives, measure efforts, create a receptive culture, and navigate organizational dynamics.
An ML System for Radiotherapy Cancer Care - From Lab to Lives
Explore ML algorithm deployment for prostate cancer radiation therapy planning in a clinical study. Learn about evaluation methods, physician assessments, and real-world impact on patient treatment.
Social Engineering with AI: From Traditional Attacks to Next-Generation Techniques
Discover how AI tools transform social engineering tactics, from crafting sophisticated phishing to conducting vishing operations, while learning both offensive techniques and defensive strategies for modern security awareness.
The Leverage of Language, or, How I Realized Information Theory Could Save Information Security
Explore how information theory can revolutionize cybersecurity through language leverage, visualization, and innovative approaches to incident response and threat intelligence.
Discovering Dark Matter Towards Better Android Malware Heuristics
Explore advanced techniques for detecting Android malware using innovative heuristics and data analysis methods to uncover hidden threats and improve mobile security.
The Wire on Fire - The Spies Who Loved Telcos
Explore telecom espionage tactics and vulnerabilities in this IT security talk, uncovering threats to communication networks and discussing protective measures.
Does Spam Have an Opinion
Explore sentiment analysis of spam messages using cloud computing and machine learning, with a humorous twist involving llama photos.
The Modern Trolley Problem
Exploring ethical dilemmas in self-driving car algorithms, demonstrating object recognition, and discussing responsible AI principles and sustainable engineering practices.
Ethical Machine Learning
Explore ethical considerations in machine learning development, focusing on mitigating bias and implementing governance from the start to ensure responsible AI deployment in enterprise settings.
Making Every Game and App More Accessible with .NET and AI
Explore Project Enabled Play: a .NET platform enabling diverse inputs for gaming and app control. Learn about scaling, accessibility, and integrating AI technologies across multiple platforms.
The Modern Trolley Problem - Responsible AI Principles
Exploring ethical challenges in AI-driven vehicles, from object recognition to decision-making algorithms. Demonstrates model training and deployment while emphasizing responsible AI principles and sustainable engineering practices.
Reducing the Environmental Footprint in Nautical Transport with F# & Serverless
Explore how F# and serverless tech optimize fuel efficiency in nautical transport, reducing carbon footprint through AI-driven predictions and cloud-based solutions.