https://www.swebb.io/blog daily 1.0 2025-12-07 https://www.swebb.io/blog/when-nobody-picks-up monthly 0.5 2024-07-25 https://images.squarespace-cdn.com/content/v1/5f78cfd6b9a6a855a40d526f/a74aea20-a615-4af0-975b-e40c2cfa171f/polling_bias.jpeg Blog - When Nobody Picks Up… - Make it stand out Whatever it is, the way you tell your story online can make all the difference. https://images.squarespace-cdn.com/content/v1/5f78cfd6b9a6a855a40d526f/20752975-498d-4d68-8e7c-0a7beece3b10/Bayes_polling.jpg Blog - When Nobody Picks Up… - Make it stand out Whatever it is, the way you tell your story online can make all the difference. https://www.swebb.io/blog/the-measure-of-a-metric monthly 0.5 2024-05-20 https://images.squarespace-cdn.com/content/v1/5f78cfd6b9a6a855a40d526f/dc2c82d8-6cdf-491a-b097-9c2ffc9fd0b0/community-donald-glover.gif Blog - The Measure of a Metric - Make it stand out Data Scientists excited to start building models firing up SQL queries. https://images.squarespace-cdn.com/content/v1/5f78cfd6b9a6a855a40d526f/cb064cfe-a76a-4094-bccd-d999805d71a4/churn.jpg Blog - The Measure of a Metric - Make it stand out Whatever it is, the way you tell your story online can make all the difference. https://www.swebb.io/blog/after-viral monthly 0.5 2024-05-07 https://images.squarespace-cdn.com/content/v1/5f78cfd6b9a6a855a40d526f/958cbfbb-03de-4d70-bedf-078db4f77b3d/postviral.jpg Blog - After Viral - Make it stand out Whatever it is, the way you tell your story online can make all the difference. https://images.squarespace-cdn.com/content/v1/5f78cfd6b9a6a855a40d526f/d6552c97-acab-4752-b8fb-1eae15333aa0/dailyrevenue.jpg Blog - After Viral - Make it stand out Whatever it is, the way you tell your story online can make all the difference. https://images.squarespace-cdn.com/content/v1/5f78cfd6b9a6a855a40d526f/d7a7a6c2-6361-4bd1-b6ab-62d3af59e879/foottraffic.jpg Blog - After Viral - Make it stand out Whatever it is, the way you tell your story online can make all the difference. https://images.squarespace-cdn.com/content/v1/5f78cfd6b9a6a855a40d526f/76f394a8-1078-48a2-8a48-29d113281abd/goneviral.jpg Blog - After Viral - Make it stand out Whatever it is, the way you tell your story online can make all the difference. https://www.swebb.io/blog/interpreting-the-categorical-cross-entropy-loss-function monthly 0.5 2024-05-09 https://images.squarespace-cdn.com/content/v1/5f78cfd6b9a6a855a40d526f/48af12f7-8fcb-4a4b-b4e4-8786f09d5410/categorical_crossentropy.png Blog - Interpreting the Categorical Cross-Entropy Loss Function - Make it stand out The categorical cross-entropy loss function, which I called S instead of L because I’m a physicist and it’s entropy. https://images.squarespace-cdn.com/content/v1/5f78cfd6b9a6a855a40d526f/3cd133bc-8ebd-44b3-8904-5cb3196c775f/entropy.png Blog - Interpreting the Categorical Cross-Entropy Loss Function - Make it stand out per-data-point categorical cross-entropy for even distribution across $n$ categories. https://www.swebb.io/blog/this-is-how-the-world-ends monthly 0.5 2023-06-22 https://www.swebb.io/blog/the-five-stages-of-a-data-science-project monthly 0.5 2022-04-14 https://www.swebb.io/blog/a-head-for-the-polls monthly 0.5 2021-06-14 https://images.squarespace-cdn.com/content/v1/5f78cfd6b9a6a855a40d526f/1622759991833-WSZ9T3QFS6HBP3KX8ECJ/training_three.png Blog - A head for the polls - Make it stand out Whatever it is, the way you tell your story online can make all the difference. https://images.squarespace-cdn.com/content/v1/5f78cfd6b9a6a855a40d526f/1622760803696-S527CC4A1NG4SSS7QPE2/senate_full_predictions.png Blog - A head for the polls - Make it stand out Whatever it is, the way you tell your story online can make all the difference. https://images.squarespace-cdn.com/content/v1/5f78cfd6b9a6a855a40d526f/1622759828739-91LUQT8HJNY3ZR8APQLF/training_one.png Blog - A head for the polls - Make it stand out Whatever it is, the way you tell your story online can make all the difference. https://images.squarespace-cdn.com/content/v1/5f78cfd6b9a6a855a40d526f/1622764235641-NUGXOECWELZ2R2JV9CV6/election_results.png Blog - A head for the polls - Make it stand out Whatever it is, the way you tell your story online can make all the difference. https://images.squarespace-cdn.com/content/v1/5f78cfd6b9a6a855a40d526f/1622680077303-8FMYYCCCURJCRCRG67NB/votes_vs_regs.png Blog - A head for the polls - Make it stand out Yep, that’s a trend https://images.squarespace-cdn.com/content/v1/5f78cfd6b9a6a855a40d526f/1622759918670-3ZGXA6WNOMDAFUEE2NFI/training_two.png Blog - A head for the polls - Make it stand out Whatever it is, the way you tell your story online can make all the difference. https://www.swebb.io/blog/stumbling-through-markov-chains monthly 0.5 2021-03-21 https://images.squarespace-cdn.com/content/v1/5f78cfd6b9a6a855a40d526f/1616270243871-SPSIM2H5RJDSJTK1KZ6Q/random_walk.png Blog - Stumbling Through Markov Chains Evolution of the distribution from initially starting on the left sidewalk. Dashed lines indicate the equilibrium probability. https://www.swebb.io/blog/bayesian-basics monthly 0.5 2021-02-02 https://images.squarespace-cdn.com/content/v1/5f78cfd6b9a6a855a40d526f/1611969837749-PFD449DCCXBFYJA56Q9M/Bayes_chart.png Blog - Bayesian Basics The branching logic of computing conditional probability https://www.swebb.io/blog/sommelier-numrique monthly 0.5 2021-02-02 https://images.squarespace-cdn.com/content/v1/5f78cfd6b9a6a855a40d526f/1611444808043-ZYC9YLO5FZNQMOW9IW03/gram_dist.png Blog - Sommelier numérique The distribution of lengths of wine vectors computed by the total number of words in the vector. https://images.squarespace-cdn.com/content/v1/5f78cfd6b9a6a855a40d526f/1611524581546-EYQRBU04NP5G6KVRYPY3/r_matrix.png Blog - Sommelier numérique Coverage of dot products for a subset of wines — each row and column is a wine, black shows a non-zero dot product between the wine vectors, white shows a zero dot product. https://www.swebb.io/blog/understanding-multi-armed-bandits monthly 0.5 2021-02-02 https://images.squarespace-cdn.com/content/v1/5f78cfd6b9a6a855a40d526f/1609713598882-HQ7CBG1MVZFYAZ1IXYJL/dropping_payrates.png Blog - Solving Multi-Armed Bandits https://images.squarespace-cdn.com/content/v1/5f78cfd6b9a6a855a40d526f/1609782414458-3EDBWFWQ735JM8RQBHXU/payout_random_dropping_eps.png Blog - Solving Multi-Armed Bandits $\varepsilon$-greedy wins the day https://images.squarespace-cdn.com/content/v1/5f78cfd6b9a6a855a40d526f/1609713475604-OZJ1E22YLMQIN7JI9JL6/payout_random_dropping.png Blog - Solving Multi-Armed Bandits Total payout comparison of dropping versus random selection https://images.squarespace-cdn.com/content/v1/5f78cfd6b9a6a855a40d526f/1609711106857-0B927KLWMRRKU4UJP4SD/random_payrates.png Blog - Solving Multi-Armed Bandits Payout and uncertainty for our ten bandit example https://images.squarespace-cdn.com/content/v1/5f78cfd6b9a6a855a40d526f/1609710970678-GJ0QH1CM5Z0LWJTJSTXH/payout_random.png Blog - Solving Multi-Armed Bandits Total payout with each pull for the random approach, plotted against the expected payout taken by averaging all the bandit payouts https://images.squarespace-cdn.com/content/v1/5f78cfd6b9a6a855a40d526f/1609716344725-9SSY4H3C8X24LY6KSJQV/greedy_payrates.png Blog - Solving Multi-Armed Bandits Computed pay rates using the $\varepsilon$-greedy strategy https://www.swebb.io/blog/a-machine-learning-quicky-why-use-l2-regularization monthly 0.5 2021-02-02 https://www.swebb.io/blog/modeling-the-market monthly 0.5 2021-02-02 https://images.squarespace-cdn.com/content/v1/5f78cfd6b9a6a855a40d526f/1603666279452-1QIAAONC9O60OAA3GU55/sp500_normed.png Blog - Modeling the Market The normalized performance of the S&P 500. https://images.squarespace-cdn.com/content/v1/5f78cfd6b9a6a855a40d526f/1603665675523-YDLC6Q70S6HBYON83V3N/exp_fit_annotated.png Blog - Modeling the Market Fitting an exponential line to the logarithmic data. https://images.squarespace-cdn.com/content/v1/5f78cfd6b9a6a855a40d526f/1607891880643-01BKU5CCZJA93ZARVUIG/mc_cdf_dist.png Blog - Modeling the Market Normalized distribution (blue) and CDF (red) https://images.squarespace-cdn.com/content/v1/5f78cfd6b9a6a855a40d526f/1607891425147-NCNWJV883X8VGDOYWZYR/monte_carlo.png Blog - Modeling the Market The many Monte Carlo simulations of our Markov Chain model https://images.squarespace-cdn.com/content/v1/5f78cfd6b9a6a855a40d526f/1604192244037-6EIF27Y65EF43TL97KM8/percent_change_fit_comparisons.png Blog - Modeling the Market Comparison of a gaussian and exponential fit to the distribution. https://images.squarespace-cdn.com/content/v1/5f78cfd6b9a6a855a40d526f/1602554516703-WSXMYDR33QHOVLB1U1K2/sp500.png Blog - Modeling the Market Historical S&P 500 data, viz’d. https://images.squarespace-cdn.com/content/v1/5f78cfd6b9a6a855a40d526f/1604193045655-S5P0UPMHCIQVXSCV6BEF/time_correlation_percent_growth.png Blog - Modeling the Market 2D histogram of percent changes today versus tomorrow for the data, which is not completely uncorrelated. The correlations day-to-day are small, but not unimportant. https://images.squarespace-cdn.com/content/v1/5f78cfd6b9a6a855a40d526f/1604633425844-LL2W57DRG0TIMYV1I5J9/loglog_annotated.png Blog - Modeling the Market A semilog plot, where the big tick markets represent a ten-fold increase. https://images.squarespace-cdn.com/content/v1/5f78cfd6b9a6a855a40d526f/1607891818244-F48MXOEESP43QIFRILH2/mc_dist.png Blog - Modeling the Market Distribution of outcomes from our Markov Chain model https://images.squarespace-cdn.com/content/v1/5f78cfd6b9a6a855a40d526f/1607891399014-ANMN156ZBPG1XCXQIA94/single_run.png Blog - Modeling the Market Example of relative performance to the trend line https://www.swebb.io/blog/programming-python-peak-performance monthly 0.5 2021-02-02 https://www.swebb.io/blog/category/Education monthly 0.5 https://www.swebb.io/blog/category/Bayes%27+Theorem monthly 0.5 https://www.swebb.io/blog/category/Probability monthly 0.5 https://www.swebb.io/blog/category/Projects monthly 0.5 https://www.swebb.io/blog/tag/numba monthly 0.5 https://www.swebb.io/blog/tag/High+Performance+Computing monthly 0.5 https://www.swebb.io/blog/tag/path+integral monthly 0.5 https://www.swebb.io/blog/tag/forecasting monthly 0.5 https://www.swebb.io/blog/tag/Bayes%27+Theorem monthly 0.5 https://www.swebb.io/blog/tag/regularization monthly 0.5 https://www.swebb.io/blog/tag/Probability monthly 0.5 https://www.swebb.io/blog/tag/loss+functions monthly 0.5 https://www.swebb.io/blog/tag/A%2FB+testing monthly 0.5 https://www.swebb.io/blog/tag/machine+learning monthly 0.5 https://www.swebb.io/blog/tag/monte+carlo monthly 0.5 https://www.swebb.io/blog/tag/elections monthly 0.5 https://www.swebb.io/blog/tag/markov+chain monthly 0.5 https://www.swebb.io/blog/tag/pandas monthly 0.5 https://www.swebb.io/blog/tag/random+walk monthly 0.5 https://www.swebb.io/blog/tag/recommender+systems monthly 0.5 https://www.swebb.io/blog/tag/small+data monthly 0.5 https://www.swebb.io/blog/tag/metropolis-hastings monthly 0.5 https://www.swebb.io/blog/tag/optimization monthly 0.5 https://www.swebb.io/blog/tag/hdf5 monthly 0.5 https://www.swebb.io/blog/tag/mutli-armed+bandit monthly 0.5 https://www.swebb.io/blog/tag/bag-of-words monthly 0.5 https://www.swebb.io/blog/tag/Python monthly 0.5 https://www.swebb.io/blog/tag/numexpr monthly 0.5 https://www.swebb.io/about daily 0.75 2020-12-08 https://images.squarespace-cdn.com/content/v1/5f78cfd6b9a6a855a40d526f/1603853889672-Z31YAWAXP5F3L9K1AH81/stephen_profile.jpg About https://www.swebb.io/contact daily 0.75 2020-10-28