My research contributions span machine learning, natural language processing, speech recognition, human-computer interaction, and distributed systems. I am not actively publishing new research, but I am always open to collaboration.
While not actively performing academic research, I am interested in the intersection of AI/ML with finance, private equity, software engineering and the overall impact on the entire entrepreneurship ecosystem.
Michael Brenndoerfer
In Progress • 2025
A comprehensive guide covering the mathematical foundations and practical implementations of machine learning, optimization, and artificial intelligence. From fundamental concepts to advanced techniques, this handbook provides both theoretical depth and real-world applications for data scientists, ML engineers, researchers, and students.
Read online
Michael Brenndoerfer
In Progress • 2025
A comprehensive guide to language AI that bridges the gap between fundamental concepts and cutting-edge applications. From the early symbolic foundations to modern transformer architectures, this book provides practical insights for practitioners working with language models, NLP systems, and AI applications.
Read onlineErik Edwards, Michael Brenndoerfer, Amanda Robinson, Najmeh Sadoughi, Gregory Finley, Maxim Korenevsky, Nico Axtmann, Mark Miller, David Suendermann-Oeft
SPECOM 2018 (20th International Conference on Speech and Computer) • 2018
Gregory Finley, Erik Edwards, Wael Salloum, Amanda Robinson, Najmeh Sadoughi, Nico Axtmann, Maxim Korenevsky, Michael Brenndoerfer, Mark Miller, David Suendermann-Oeft
SPECOM 2018 (20th International Conference on Speech and Computer) • 2018
Najmeh Sadoughi, Gregory Finley, Erik Edwards, Amanda Robinson, Maxim Korenevsky, Michael Brenndoerfer, Nico Axtmann, Mark Miller, David Suendermann-Oeft
SPECOM 2018 (20th International Conference on Speech and Computer) • 2018
Gregory Finley, Erik Edwards, Amanda Robinson, Najmeh Sadoughi, James Fone, Mark Miller, David Suendermann-Oeft, Michael Brenndoerfer, Nico Axtmann
INTERSPEECH 2018 • 2018
Gregory Finley, Wael Salloum, Najmeh Sadoughi, Erik Edwards, Amanda Robinson, Nico Axtmann, Michael Brenndoerfer, Mark Miller, David Suendermann-Oeft
NAACL-HLT 2018 (Industry Track) • 2018
Gregory Finley, Erik Edwards, Amanda Robinson, Michael Brenndoerfer, Najmeh Sadoughi, James Fone, Nico Axtmann, Mark Miller, David Suendermann-Oeft
NAACL-HLT 2018 (Demonstrations) • 2018
Casey Dugan, Aabhas Sharma, Michael Muller, Di Lu, Michael Brenndoerfer, Werner Geyer
INTERACT 2017 (IFIP Conference on Human-Computer Interaction) • 2017
Michael Muller, Casey Dugan, Michael Brenndoerfer, Megan Monroe, Werner Geyer
CSCW 2017 (ACM Conference on Computer-Supported Cooperative Work and Social Computing) • 2017
Michael Brenndoerfer
LinkedIn Article • 2021
Michael Brenndoerfer
LinkedIn Article • 2021
Michael Brenndoerfer, Stefan Palombo, Vinitra Swamy
Distill Literature Review • 2018
Michael Brenndoerfer, by Jessie Ying
Berkeley Master of Engineering (Medium) • 2018
Michael Brenndoerfer, edited by Maya Rector
Berkeley Master of Engineering (Medium) • 2018
In the 1980s, neural networks hit a wall—nobody knew how to train deep models. That changed when Rumelhart, Hinton, and Williams introduced backpropagation in 1986. Their clever use of the chain rule finally let researchers figure out which parts of a network deserved credit or blame, making deep learning work in practice. Thanks to this breakthrough, we now have everything from word embeddings to powerful language models like transformers.
In 2002, IBM researchers introduced BLEU (Bilingual Evaluation Understudy), revolutionizing machine translation evaluation by providing the first widely adopted automatic metric that correlated well with human judgments. By comparing n-gram overlap with reference translations and adding a brevity penalty, BLEU enabled rapid iteration and development, establishing automatic evaluation as a fundamental principle across all language AI.
In 1988, Yann LeCun introduced Convolutional Neural Networks at Bell Labs, forever changing how machines process visual information. While initially designed for computer vision, CNNs introduced automatic feature learning, translation invariance, and parameter sharing. These principles would later revolutionize language AI, inspiring text CNNs, 1D convolutions for sequential data, and even attention mechanisms in transformers.
In 2001, Lafferty and colleagues introduced CRFs, a powerful probabilistic framework that revolutionized structured prediction by modeling entire sequences jointly rather than making independent predictions. By capturing dependencies between adjacent elements through conditional probability and feature functions, CRFs became essential for part-of-speech tagging, named entity recognition, and established principles that would influence all future sequence models.
Joseph Weizenbaum's ELIZA, created in 1966, became the first computer program to hold something resembling a conversation. Using clever pattern-matching techniques, its famous DOCTOR script simulated a Rogerian psychotherapist. ELIZA showed that even simple tricks could create the illusion of understanding, bridging theory and practice in language AI.
In 1987, Slava Katz solved one of statistical language modeling's biggest problems. When your model encounters word sequences it has never seen before, what do you do? His elegant solution was to "back off" to shorter sequences, a technique that made n-gram models practical for real-world applications. By redistributing probability mass and using shorter contexts when longer ones lack data, Katz back-off allowed language models to handle the infinite variety of human language with finite training data.
In 1997, Hochreiter and Schmidhuber solved the vanishing gradient problem with LSTMs, introducing sophisticated gated memory mechanisms that could selectively remember and forget information across long sequences. This breakthrough enabled practical language modeling, machine translation, and speech recognition while establishing principles of gated information flow that would influence all future sequence models.
Bernard Widrow and Marcian Hoff built MADALINE at Stanford in 1962, taking neural networks beyond the perceptron's limitations. This adaptive architecture could tackle real-world engineering problems in signal processing and pattern recognition, proving that neural networks weren't just theoretical curiosities but practical tools for solving complex problems.
In 1957, Frank Rosenblatt created the perceptron at Cornell University, the first artificial neural network that could actually learn to classify patterns. This groundbreaking algorithm proved that machines could learn from examples, not just follow rigid rules. It established the foundation for modern deep learning and every neural network we use today.
In 1995, RNNs revolutionized sequence processing by introducing neural networks with memory—connections that loop back on themselves, allowing machines to process information that unfolds over time. This breakthrough enabled speech recognition, language modeling, and established the sequential processing paradigm that would influence LSTMs, GRUs, and eventually transformers.
Claude Shannon's 1948 work on information theory introduced n-gram models, one of the most foundational concepts in natural language processing. These deceptively simple statistical models predict language patterns by looking at sequences of words. They laid the groundwork for everything from autocomplete to machine translation in modern language AI.
Terry Winograd's 1968 SHRDLU system took a revolutionary approach to language understanding. Instead of just pattern matching, it created genuine comprehension within a simulated blocks world. SHRDLU could parse complex sentences, track what was happening, and demonstrate that computers could truly understand language when grounded in a physical context.
In 1991, IBM researchers revolutionized machine translation by introducing the first comprehensive statistical approach. Instead of hand-crafted linguistic rules, they treated translation as a statistical problem of finding word correspondences from parallel text data. This breakthrough established principles like data-driven learning, probabilistic modeling, and word alignment that would transform not just translation, but all of natural language processing.
Natural language processing underwent a fundamental shift from symbolic rules to statistical learning. Early systems relied on hand-crafted grammars and formal linguistic theories, but their limitations became clear. The statistical revolution of the 1980s transformed language AI by letting computers learn patterns from data instead of following rigid rules.
In 1987, Alex Waibel introduced Time Delay Neural Networks, a revolutionary architecture that changed how neural networks process sequential data. By introducing weight sharing across time and temporal convolutions, TDNNs laid the groundwork for modern convolutional and recurrent networks. This breakthrough enabled end-to-end learning for speech recognition and established principles that remain fundamental to language AI today.
In 1950, Alan Turing proposed a deceptively simple test for machine intelligence, originally called the Imitation Game. Could a machine fool a human judge into thinking it was human through conversation alone? This thought experiment shaped decades of AI research and remains surprisingly relevant today as we evaluate modern language models like GPT-4 and Claude.
In 1995, Princeton University released WordNet 1.0, a revolutionary lexical database that represented words not as isolated definitions, but as interconnected concepts in a semantic network. By capturing relationships like synonymy, hypernymy, and meronymy, WordNet established the principle that meaning is relational, influencing everything from word sense disambiguation to modern word embeddings and knowledge graphs.
Learn about multicollinearity in regression analysis with this practical guide. VIF analysis, correlation matrices, coefficient stability testing, and approaches such as Ridge regression, Lasso, and PCR. Includes Python code examples, visualizations, and useful techniques for working with correlated predictors in machine learning models.
A comprehensive guide to Ordinary Least Squares (OLS) regression, including mathematical derivations, matrix formulations, step-by-step examples, and Python implementation. Learn the theory behind OLS, understand the normal equations, and implement OLS from scratch using NumPy and scikit-learn.
A complete hands-on guide to simple linear regression, including formulas, intuitive explanations, worked examples, and Python code. Learn how to fit, interpret, and evaluate a simple linear regression model from scratch.
A comprehensive guide to R-squared, the coefficient of determination. Learn what R-squared means, how to calculate it, interpret its value, and use it to evaluate regression models. Includes formulas, intuitive explanations, practical guidelines, and visualizations.
Learn how to build agentic workflows with LangChain and LangGraph.
An in-depth look at what happens to money during market crashes, how wealth is redistributed, and the mechanisms behind market recovery.
Understand the mathematical foundations of LLM fine-tuning with clear explanations and minimal prerequisites. Learn how gradient descent, weight updates, and Transformer architectures work together to adapt pre-trained models to new tasks.
A comprehensive guide to choosing the right approach for your LLM project: using pre-trained models as-is, enhancing them with context injection and RAG, or specializing them through fine-tuning. Learn the trade-offs, costs, and when each method works best.
Learn the foundational concepts of LLM workflows - connecting language models to tools, handling responses, and building intelligent systems that take real-world actions.
Learn how to use Monte Carlo simulation to model and analyze stock market returns, estimate future performance, and understand the impact of randomness in financial forecasting. This tutorial covers the fundamentals, practical implementation, and interpretation of simulation results.
A comprehensive guide to understanding AI agents, their building blocks, and how they differ from agentic workflows and agent swarms.
A deep dive into how MCP makes tool use with LLMs easier, cleaner, and more standardized.
An exploration of why setting temperature to zero doesn't eliminate all randomness in large language model outputs.
Get notified when I publish new articles on data and AI, private equity, technology, and more.