Skip to content
View stochastic-sisyphus's full-sized avatar
🏔️
Hiking
🏔️
Hiking
7 followers · 30 following

Achievements

Achievement: YOLOAchievement: Pull Shark

Achievements

Achievement: YOLOAchievement: Pull Shark

Highlights

Block or report stochastic-sisyphus

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Please don't include any personal information such as legal names or email addresses. Maximum 100 characters, markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
stochastic-sisyphus/README.md

Welcome to the mountain. I'm Sisyphus, and this is my boulder.

Well, to be precise—I'm him with an adjective slapped in front: Stochastic Sisyphus™. Unless we're in reality, where I have a real name, allegedly.

Vanessa B, MSBA

But that's just semantics

banner img or whatever

Everything In One Place self.define() GitHub Portfolio GitHub Research ORCID ID

Let's start with the obligatory passion statement, because that seems necessary.

My process is a perpetual cycle of curiosity, transforming complexity into understanding and expanding the limits of knowledge. Learning—machine, deep, and human

Honestly, the fact that I named my cat Marvin Minsky is a better passion statement than any string of words I can come up with

The machine is learning, under my supervision (of course). Ironically, I remain confused. Recursion: the cause and solution to all my problems.

The Sisyphean struggle means rolling the boulder in search of answers at the base, getting lost in thought, then pushing algorithms up the mountain with each iteration. As a masochist, I spend my time optimizing models that refuse to cooperate—because apparently, I find trial and error entertaining.

This is a snapshot of ongoing work and the evolution of my projects over time. Most code gets rewritten, some ideas never make it past their first push—an ongoing hypothesis test proving that not every idea deserves a second look.

Typically Focused On:

High confidence, very predictable

  • Machine Learning: Transformers, semantic search, multilingual modeling, contrastive learning, active learning, probabilistic models, causal inference, reinforcement learning, and algorithms that learn and some that refuse to behave
  • NLP: Retrieval, embeddings, language models, making models understand nuance (or at least fake it)
  • Uncertainty & Complexity: Finding patterns in noise, making sense of what isn't obvious, modeling causal relationships

Made Possible by Viewers Like You:

Not a fixed stack, just the things I use (or break) the most, a special shoutout to Python

  • Frameworks: Python, PyTorch, TensorFlow, Keras, Scikit-Learn, XGBoost
  • Systems: Cloud pipelines, distributed processing, deployment, Docker, GPUs

The Cycle

If nothing else, the name is accurate. Just me, the boulder, and the incline.

           Perpetually uphill,

           Stochastic Sisyphus

Pinned Loading

  1. Portfolio Portfolio Public

    This repository serves as a collection of my coding projects over the years & a snapshot of ongoing work. Projects at various stages of completion, experiments that may lead somewhere (or not), and…

    Jupyter Notebook 1

  2. research research Public

    My process is a perpetual cycle of curiosity, transforming complexity into understanding and expanding the limits of knowledge. Learning—machine, deep, and human.

    1

  3. adv_data_processing_pipeline adv_data_processing_pipeline Public

    Python 1

  4. prophetic-emergentomics prophetic-emergentomics Public

    The Prophecy of the Emergent Economy: Beyond Prediction—Frameworks for Economic Evolution

    1

  5. synsearch synsearch Public

    Python 1

  6. feature-selection-optuna-remix feature-selection-optuna-remix Public

    Python 1