👋
Hi! My name is Edwin, and I'm currently a data scientist at Nibble looking at ways of using machine learning to improve a conversational AI agent for negotiation. I recently finished my MSc in Statistics with Data Science at the School of Mathematics, University of Edinburgh.
I normally work with Python, R and Ruby, mainly doing machine learning or data science related things in Python and R, and any general purpose scripting, task automation or web development in Ruby – but I'm always interested in learning new things!
Right now, I'm learning about:
- Gaussian processes
- Data engineering
I'd like to learn more about:
- Bayesian ML: variational inference, probablistic graphical models (e.g. CRFs)
- Statistical time series concepts: autocorrelation, forecasting models (ARIMA, GARCH etc.)
- Signal processing for ML: STFT, Mel spectrograms, MFCCs and chromagrams
- State-of-the-art NLP language modelling techniques: transformers, BERT and GPT-3
- Ensemble classifiers: bagging and boosting (with AdaBoost, XGBoost, LightGBM etc.)
I'm very familiar with:
- Common ML methods: GLM, logistic regression, kNN, mixture models etc.
- Statistical methodology: likelihood-based inference (MLEs, confidence intervals), hypothesis testing
- Neural networks: mainly feed-forward and recurrent architectures, but also some knowledge and practice with CNNs
- Sequence classification algorithms: HMMs, RNNs (LSTM/GRU), DTW + kNN
- Natural language processing: attention, n-gram language models, POS tagging, word embeddings, sentiment analysis
I am confident with these languages, tools and systems:
|
|
|
|
|
|
|
|---|---|---|---|---|---|---|
|
|
|
|
|
|
|