How to Ground LLM’s to minimize hallucinations

Language models have made significant advancements in recent years, with models like GPT-3 and GPT-4 showcasing impressive capabilities. However, one persistent challenge that arises with these models is the occurrence of hallucinations—instances where the model generates plausible-sounding but incorrect or nonsensical responses.

In this talk, we will explore strategies to ground language models to minimize the occurrence of hallucinations. By grounding LLMs, we aim to enhance their reliability and ensure that the generated outputs align more closely with factual accuracy and logical coherence.

We will discuss various techniques and approaches that can be employed to address hallucinations effectively. These may include fine-tuning the models on domain-specific data, incorporating external knowledge sources, leveraging human-in-the-loop feedback, and implementing robust evaluation mechanisms.

Furthermore, we will delve into the underlying causes of hallucinations and examine the limitations of current language models. By understanding these factors, we can develop targeted strategies to mitigate the occurrence of hallucinations and improve the overall performance of LLMs.

Join us in this talk as we explore practical methods to ground language models and minimize hallucinations. Discover how these techniques can enhance the reliability and trustworthiness of LLMs, making them more suitable for real-world applications across various domains. Together, let’s unlock the full potential of language models while ensuring their outputs align with factual accuracy and logical coherence.

Morality and AI: How do I use this power responsibly?

As artificial intelligence (AI) continues to advance, it becomes increasingly important to address the ethical implications and moral considerations surrounding its use. In this talk, we will explore the intersection of morality and AI, focusing on how individuals and organizations can wield this power responsibly.

AI, including language models like GPT-3 and GPT-4, possesses immense potential to shape our society and influence decision-making processes. However, with great power comes great responsibility. We will delve into the ethical challenges and moral dilemmas that arise when utilizing AI, particularly in the context of language models.

Join us as we discuss the principles and frameworks that can guide responsible AI usage. We will explore topics such as fairness, transparency, accountability, and privacy. By understanding these principles, individuals and organizations can ensure that their AI applications align with ethical standards and societal values.

Furthermore, we will examine the importance of diverse and inclusive data in training AI models. We will address the risks of bias and discrimination and discuss strategies to mitigate these issues, promoting fairness and equity in AI systems.

Through real-world case studies and examples, we will highlight the potential consequences of irresponsible AI use and the importance of proactive ethical considerations.

Join us in this talk to gain a deeper understanding of the moral dimensions of AI and how to navigate them responsibly. Discover how to leverage AI, including language models, in a manner that aligns with ethical principles and contributes positively to society. Together, let’s harness the power of AI while upholding our moral obligations.

Creating your own ML powered Holographic Assistant Workshop

I’ve been fascinated with Mixed Reality since the introduction of the HoloLens.  There is something magical about combining Artificial Intelligence with Holograms.  In this workshop, I will show you a practical example of how to mix holograms and AI to create your own holographic assistant.

For this workshop, I will skip the theory and go directly to application using off-the-shelf tooling and models.  I hope to inspire the audience to look closer at the capabilities of off-the-shelf machine learning by tapping into the power of ML without building any custom models.

This talk will make use of a lot of different tools and technologies all available, accessible, and inexpensive.  I will use C#, Unity, a Holographic Display, Azure Cognitive Services, and Azure Open AI to create an interactive experience pushing the envelope of current tools and techniques.

Each attendee will create their own assistant using their own laptop, Azure subscription, Visual Studio, and Unity.  At the end of the workshop attendees will have the opportunity to demonstrate their assistant on a holographic display but will also build a working assistant that runs on 2D displays.

  • See an end-to-end solution created around LLM’s
  • Inspire the attendee to do more than create another text based chat bot
  • Help the attendee to walk away with their own working assistant

Recording of AI Tech Shop Panel Available

The recording of the AI Tech Shop Panel is available here:

Thanks to the organizers and other panelists.  This was a great panel with interesting content from each of the presenters.  Check out the recording and feel free to reach out with questions!

Tech-Prize Artificial Intelligence Panel

Join me this Saturday at the Tech Shop AI Panel Discussion.   This panel will include panelists from SE Wisconsin business and education with expertise in Artificial Intelligence, Machine Learning, and its application in app development and decision-making systems. The panel will overview AI/ML technologies and discuss the process, techniques, and technology used to create apps. The panel will also discuss how our institutions are keeping pace with AI, discuss career opportunities and the skills needed to enter the field.

Sign up here:

Artificial Intelligence Tech Shop



MACC 2021 Why do Machine Learning Projects Fail Slides

It was great to return to the Midwest Architecture Community Collaboration Conference this year even if it was virtual.  This talk has been something I’ve building towards for years, and attempts to capture some of the wisdom I have acquired in the last few years in machine learning space.  ML project failures is the hidden secret nobody wants to talk about in our industry.  I hope everyone enjoyed hearing this talk as my as I enjoyed giving it!

Read More

ML .Net talk Slides

Thank you to Mad Dot  Net for hosting my talk on ML .Net.  ML .Net is exciting technology for C# developers that would like to apply machine learning to the project.  Due to my animations some of the slides didn’t translate well to the web, but all of the content is there.  The deck is posted below. Let me know if you have any questions!

Read More

Learn more about Machine Learning for C# and .NET developers

This week I’m speaking at MadDotNet about ML .NET.  ML .NET has matured since its initial release and now has a solid feature set allowing many common machine learning scenarios to be accomplished completely in C#.  This talk will be mostly focused on code with multiple working demos using real world use cases.

Sign Up Here:

Read More

Sign Ups open for September CNUG Talk

At this month’s Chicago .NET Users Group, I will be giving a talk discussing how to bring Machine Learning to your development teams in a sensible and pragmatic way.  This is a virtual online event, so I would like to encourage everyone to join us regardless of location!

Read More

Speaking tomorrow at Chicago Code Camp

Come join panelists Cameron Vetter, Greg Levenhagen, Svetlana Levitan and Swaminathan Venkatesh to discuss the latest trends in AI & ML. We’ll be discussing where AI & ML will take humanity to the future, as well as topics involving Ethics, Economics and the AI & ML Technologies themselves.

Will the robots replace our jobs in the future? The topic of AI & ML is not all about doom and gloom: how can we use AI & ML to improve our lives? Are Siri, Google, and Alexa the epitome of AI?

Join us for an hour (or maybe more?) with questions and topics you want to discuss with our Panel and the Midwest Development Community.

Please tell your friends and colleagues, and we hope to see you at 4 PM on Saturday, Saturday, July 11, 2020.

Signup here: