Learning from Machine Learning

On this episode of Learning from Machine Learning, I had the privilege of speaking with Aman Khan, Head of Product at Arize AI. Aman shared how evaluating AI systems isn't just a step in the process—it's a machine learning challenge in of itself. Drawing powerful analogies between mechanical engineering and AI, he explained, "Instead of tolerances in manufacturing, you're designing for non-determinism," reminding us that complexity often breeds opportunity. Aman's journey from self-driving cars to ML evaluation tools highlights the critical importance of robust systems that can handle failure. He encourages teams to clearly define outcomes, break down complex systems, and build evaluations into every step of the development pipeline. Most importantly, Aman's insights remind us that machine learning—much like life—is less deterministic and more probabilistic, encouraging us to question how we deal with the uncertainty in our own lives. Thank you for listening. Be sure to subscribe and share with a friend or colleague . Until next time... keep on learning.

In this episode of Learning from Machine Learning, we explore the intersection of pure mathematics and modern data science with Leland McInnes, the mind behind an ecosystem of tools for unsupervised learning including UMAP, HDBSCAN, PyNN Descent and DataMapPlot. As a researcher at the Tutte Institute for Mathematics and Computing, Leland has fundamentally shaped how we approach and understand complex data.Leland views data through a unique geometric lens, drawing from his background in algebraic topology to uncover hidden patterns and relationships within complex datasets. This perspective led to the creation of UMAP, a breakthrough in dimensionality reduction that preserves both local and global data structure to allow for incredible visualizations and clustering. Similarly, his clustering algorithm HDBSCAN tackles the messy reality of real-world data, handling varying densities and noise with remarkable effectiveness.But perhaps what's most striking about Leland isn't just his technical achievements – it's his philosophy toward algorithm development. He champions the concept of "decomposing black box algorithms," advocating for transparency and understanding over blind implementation. By breaking down complex algorithms into their fundamental components, Leland argues, we gain the power to adapt and innovate rather than simply consume.For those entering the field, Leland offers poignant advice: resist the urge to chase the hype. Instead, find your unique angle, even if it seems unconventional. His own journey – applying concepts from algebraic topology and fuzzy simplicial sets to data science – demonstrates how breakthrough innovations often emerge from unexpected connections.Throughout our conversation, Leland's passion for knowledge and commitment to understanding shine through. His approach reminds us that the most powerful advances in data science often come not from following the crowd, but from diving deep into fundamentals and drawing connections across disciplines.There's immense value in understanding the tools you use, questioning established approaches, and bringing your unique perspective to the field. As Leland shows us, sometimes the most significant breakthroughs come from seeing familiar problems through a new lens.Resources for Leland McInnesLeland’s GithubUMAPHDBSCANPyNN DescentDataMapPlotEVoCReferencesMaarten GrootendorstLearning from Machine Learning Episode 1Vincent Warmerdam - CalmcodeLearning from Machine Learning Episode 2Matt RocklinEmily Riehl - Category Theory in ContextLorena BarbaDavid Spivak - Fuzzy Simplicial SetsImproving Mapper’s Robustness by Varying Resolution According to Lens-Space DensityLearning from Machine LearningYoutubehttps://mindfulmachines.substack.com/

In this episode, we are joined by Chris Van Pelt, co-founder of Weights & Biases and Figure Eight/CrowdFlower. Chris has played a pivotal role in the development of MLOps platforms and has dedicated the last two decades to refining ML workflows and making machine learning more accessible.Throughout the conversation, Chris provides valuable insights into the current state of the industry. He emphasizes the significance of Weights & Biases as a powerful developer tool, empowering ML engineers to navigate through the complexities of experimentation, data visualization, and model improvement. His candid reflections on the challenges in evaluating ML models and addressing the gap between AI hype and reality offer a profound understanding of the field's intricacies.Drawing from his entrepreneurial experience co-founding two machine learning companies, Chris leaves us with lessons in resilience, innovation, and a deep appreciation for the human dimension within the tech landscape. As a Weights & Biases user for five years, witnessing both the tool and the company's growth, it was a genuine honor to host Chris on the show.References and Resourceshttps://wandb.ai/https://www.youtube.com/c/WeightsBiaseshttps://x.com/weights_biaseshttps://www.linkedin.com/company/wandb/https://twitter.com/vanpeltResources to learn more about Learning from Machine Learninghttps://www.youtube.com/@learningfrommachinelearninghttps://www.linkedin.com/company/learning-from-machine-learninghttps://mindfulmachines.substack.com/https://www.linkedin.com/in/sethplevine/https://medium.com/@levine.seth.p

This episode features Dr. Michelle Gill, Tech Lead and Applied Research Manager at NVIDIA, working on transformative projects like BioNemo to accelerate drug discovery through AI. Her team explores Biofoundation models to enable researchers to better perform tasks like protein folding and small molecule binding.Michelle shares her incredible journey from wet lab biochemist to driving cutting edge AI at NVIDIA. Michelle discusses the overlap and differences between NLP and AI in biology. She outlines the critical need for better machine learning representations that capture the intricate dynamics of biology.Michelle provides advice for beginners and early career professionals in the field of machine learning, emphasizing the importance of continuous learning and staying up to date with the latest tools and techniques. She also shares insights on building successful multidisciplinary teamsAfter hearing her fascinating PyData NYC keynote, it was such an honor to have her on the show to discuss innovations at the intersection of biochemistry and AI.References and Resourceshttps://michellelynngill.com/Michelle Gill - Keynote - PyData NYC https://www.youtube.com/watch?v=ATo2SzA1Pp4AlexNetAlphaFold - https://www.nature.com/articles/s41586-021-03819-2OpenFold - https://www.biorxiv.org/content/10.1101/2022.11.20.517210v1BioNemo - https://www.nvidia.com/en-us/clara/bionemo/NeurIPS - https://nips.cc/Art Palmer - https://www.biochem.cuimc.columbia.edu/profile/arthur-g-palmer-iii-phdPatrick Loria - https://chem.yale.edu/faculty/j-patrick-loriaScott Strobel - https://chem.yale.edu/faculty/scott-strobelAlexander Rives - https://www.forbes.com/sites/kenrickcai/2023/08/25/evolutionaryscale-ai-biotech-startup-meta-researchers-funding/?sh=648f1a1140cfDeborah Marks - https://sysbio.med.harvard.edu/debora-marksResources to learn more about Learning from Machine Learninghttps://www.linkedin.com/company/learning-from-machine-learninghttps://mindfulmachines.substack.com/https://www.linkedin.com/in/sethplevine/https://medium.com/@levine.seth.p

Ines Montani, Co-founder and CEO of Explosion, discusses the evolution of the web and machine learning, development of SpaCy, NLP vs. NLU, misconceptions of starting a software company, value of understanding business problems, labeling data, combining large models with specific models, evolution of Spacey and its goals, T-shaped vs Tree-shaped skills in software engineering, creating holiday special emojis as a data scientist, embracing an entrepreneurial spirit.

Lewis Tunstall, machine learning engineer at Hugging Face and author of the best selling book Natural Language Processing with Transformers, talks about his journey from theoretical physics to machine learning, the benefits of the Hugging Face platform, exploring reinforcement learning and the TRL library, adapters in fine-tuning pre-trained transformers, the limitations of language models, and understanding biases in machine learning and the impact on society.

Paige Bailey, lead product manager for generative models at Google DeepMind, discusses her work with machine learning techniques and tools, including the development of large language models like Bard. She explores the benefits and capabilities of Palm V2, the upgraded model for BARD in code, math, reasoning, and multilingual tasks. The chapter also delves into the trade-offs and accessibility of large language models, the exciting use cases of multimodal models, and the impact and importance of large language models in the field of machine learning.

Sebastian Raschka, Lead AI Educator at Lightning and author, discusses learning ML, responsible AI, AGI, and more. Topics include his career background, working in academia vs industry, early ML projects, creating a book with Chat GPT, dangers of overconfidence in ML models, balancing learning from scratch with existing solutions, and the importance of having fun and being patient.

This episode welcomes Nils Reimers, Director of Machine Learning at Cohere and former research at Hugging Face, to discuss Natural Language Processing, Sentence Transformers and the future of Machine Learning. Nils is best known as the creator of Sentence Transformers, a powerful framework for generating high-quality sentence embeddings that has become increasingly popular in the ML community with over 9K stars on Github. With Sentence Transformers, Nils has enabled researchers and developers (including me) to train state-of-the-art models for a wide range of NLP tasks, including text classification, semantic similarity, and question-answering. His contributions have been recognized by numerous awards and publications in top-tier conferences and journals.Resources to learn more about Nils Reimers and his work:https://www.nils-reimers.de/https://www.sbert.net/https://scholar.google.com/citations?...https://cohere.ai/Resources to learn more about Learning from Machine Learning:https://www.linkedin.com/company/learning-from-machine-learninghttps://www.linkedin.com/in/sethplevine/https://medium.com/@levine.seth.pYoutube Clips02:29 What attracted you to Machine Learning?06:32 What is sentence transformers?28:02 Benchmarks and P-Hacking33:53 What’s an important question that remains unanswered in Machine Learning?38:41 How do you view the gap between the hype and the reality in Machine Learning?50:45 What advice would you give to someone just starting out?52:30 What advice would you give yourself when you were just starting out in your career?57:22 What has a career in ML taught you about life?

Vincent Warmerdam, creator of calmcode and machine learning engineer at SpaCy, discusses his career path and the role of luck and privilege. They talk about different job titles and the creation of Com Code, a free platform for data science education. Vincent's open-source projects and the importance of Python packages are highlighted. They discuss reframing problems, combining ML models with heuristics, and finding inspiration in unexpected places.