[

{"title":"La mente che conta fino a dieci","url":"/sections/science/2026/06/01/The-mind-that-counts-to-ten.html","excerpt":"\n  La cognizione è diffusa e costa poco. Un sé che vive nel tempo è raro, fragile, legato a un corpo. E una macchina che rimescola l’ordine dei propri pensieri senza accorgersene è lo specchio più strano che abbiamo mai costruito.\n\n","categories":["sections","science","deep-learning"],"tags":[],"date":"2026-06-01"},

{"title":"A controlled testbed for Repeat-Your-Self — SAT solving, message passing, and a predictive rho/phi theory","url":"/sections/science/2026/05/29/RYS-controlled-SAT-message-passing-and-rho-phi-theory.html","excerpt":"RYS Controlled SAT message passing\n","categories":["sections","science","deep-learning"],"tags":[],"date":"2026-05-29"},

{"title":"Deep Equilibrium Models as a model for RYS","url":"/sections/science/2026/05/18/Deep-Equilibrium-Models-As-A-Model-For-RYS.html","excerpt":"In the last two posts I tried to put David Ng’s Repeat-Your-Self construction into a geometric language. In How skip connections define graphs in deep networks I started from the residual recursion\n","categories":["sections","science","deep-learning"],"tags":[],"date":"2026-05-18"},

{"title":"Fractal Connectivity Scaffold Networks","url":"/sections/science/2026/05/16/Fractal-connectivity-scaffold-networks.html","excerpt":"1. Question\n","categories":["sections","science"],"tags":[],"date":"2026-05-16"},

{"title":"Hippocampal scaffold networks","url":"/sections/science/2026/05/15/Hippocampal-scaffold-networks.html","excerpt":"The tradeoff\n","categories":["sections","science"],"tags":[],"date":"2026-05-15"},

{"title":"Multiscale structured recurrence with fractal connectivity networks","url":"/sections/science/2026/05/15/Multiscale-structured-recurrence-fractal-connectivty-networks.html","excerpt":"1. Introduction\n","categories":["sections","science","deep-learning"],"tags":[],"date":"2026-05-15"},

{"title":"Similarity of neural networks representations","url":"/sections/science/2026/04/29/Similarity-of-neural-networks-representations.html","excerpt":"In yesterday’s post on How skip connections define graphs in deep networks I built an \\(L\\times L\\) functional connectome of a Transformer by taking pairwise cosine similarities of layer activations. That edge had a clean closed form in the cumulative residual force \\(\\rho_{i,...","categories":["sections","science","deep-learning"],"tags":[],"date":"2026-04-29"},

{"title":"How skip connections define graphs in deep networks","url":"/sections/science/2026/04/28/Skip-connections-and-graph-analysis.html","excerpt":"The hidden graphs behind residual connections\n","categories":["sections","science"],"tags":[],"date":"2026-04-28"},

{"title":"Functional synthetic LLM connectome analysis","url":"/sections/science/2026/04/28/Functional-synthetic-llm-connectome-analysis.html","excerpt":"Connections to network neuroscience\n","categories":["sections","science","deep-learning"],"tags":[],"date":"2026-04-28"},

{"title":"PLP Inference Audited Against Murphy","url":"/sections/science/2026/04/21/PLP-inference-theoretical-basis.html","excerpt":"Kevin Murphy’s treatment of inference on pp. 435-471 of Probabilistic Machine Learning: Advanced Topics gives us a clean gold standard for what exact and approximate inference algorithms are supposed to mean. Those pages cover the forwards-backwards algorithm for hidden Markov...","categories":["sections","science","deep-learning"],"tags":[],"date":"2026-04-21"},

{"title":"The Master Equation of a Messy World","url":"/sections/science/2026/04/16/Transformers-Mimic-Nature-Equations.html","excerpt":"Transformers mimic nature equations\n","categories":["sections","science","deep-learning"],"tags":[],"date":"2026-04-16"},

{"title":"Kolmogorov-Arnold theorem, KANs, and the `eml` operator","url":"/sections/science/2026/04/14/Kolmogorov-Arnold-Theorem-And-EML-operator.html","excerpt":"A strange echo\n","categories":["sections","science"],"tags":[],"date":"2026-04-14"},

{"title":"Amortized structural variational inference for probabilistic language programming","url":"/sections/science/2026/04/14/Agentic-soft-logical-circuits-amortized-structured-VI.html","excerpt":"I have been spending a lot of time lately staring at agent logs.\n","categories":["sections","science","deep-learning"],"tags":[],"date":"2026-04-14"},

{"title":"The continuous Sheffer stroke is all you need","url":"/sections/science/2026/04/13/Continuous-Sheffer-stroke-all-you-need.html","excerpt":"All elementary functions from a single operator\n","categories":["sections","science","statistical-learning"],"tags":[],"date":"2026-04-13"},

{"title":"From brain to disk: how I built a voice-activated second brain for $0","url":"/sections/tech/tech,/productivity/2026/04/11/voice-activated-second-brain.html","excerpt":"We’ve all been there. You’re out for a walk or just about to fall asleep when a perfect idea hits. You grab your phone, record a quick voice note, and then… it just sits there. It becomes a digital fossil in your chat history that you’ll probably never listen to again.\n","categories":["sections","tech","tech,","productivity"],"tags":["obsidian,","AI,","automation,","secondbrain"],"date":"2026-04-11"},

{"title":"Interpretations for the Kullback-Leibler divergence, or relative entropy","url":"/sections/science/2026/04/09/Explanations-for-KL-divergence.html","excerpt":"This page follows the structure of Six (and a half) intuitions for KL divergence by Callum McDougall (2022); equations are set for MathJax.\n","categories":["sections","science","statistical-learning"],"tags":[],"date":"2026-04-09"},

{"title":"Latent traces and Landau free energy","url":"/sections/science/2026/04/05/latent-traces-and-landau-free-energy.html","excerpt":"Latent traces, not narrated ODEs\n","categories":["sections","science","deep-learning"],"tags":[],"date":"2026-04-05"},

{"title":"Chain-of-thought as grand-canonical inference in energy-based language models","url":"/sections/science/2026/04/05/chain-of-thought-grand-canonical-inference.html","excerpt":"From zero-shot decoding to grand-canonical reasoning\n","categories":["sections","science","deep-learning"],"tags":[],"date":"2026-04-05"},

{"title":"Statistical mechanics of decision making","url":"/sections/science/2026/04/05/Statistical-mechanics-decision-making.html","excerpt":"A recurring question about language models is deceptively simple to state. When does asking a model to “think step by step” actually help it decide, and when does it just produce longer text? I want to give that question a precise answer, and the cleanest language I know for i...","categories":["sections","science","deep-learning"],"tags":[],"date":"2026-04-05"},

{"title":"Latent traces and Landau free energy","url":"/sections/science/2026/04/05/Latent-traces-and-Landau-free-energy.html","excerpt":"Latent traces, not narrated ODEs\n","categories":["sections","science","deep-learning"],"tags":[],"date":"2026-04-05"},

{"title":"Latent traces and Landau free energy","url":"/sections/science/2026/04/05/Latent-Traces-and-Landau-Free-Energy.html","excerpt":"Latent traces, not narrated ODEs\n","categories":["sections","science","deep-learning"],"tags":[],"date":"2026-04-05"},

{"title":"From hard to soft operators: between machine learning and statistical physics","url":"/sections/science/2026/04/05/From-hard-to-soft-operators-machine-learning-and-physics.html","excerpt":"An academic exploration of deep learning, statistical mechanics, and category theory\n","categories":["sections","science","language-physics"],"tags":[],"date":"2026-04-05"},

{"title":"Random rooted trees, continuation free energy, and the Diligent Learner","url":"/sections/science/2026/04/04/Random-rooted-trees-free-energy-and-the-diligent-learner.html","excerpt":"Why these two theories should meet\n","categories":["sections","science","deep-learning"],"tags":[],"date":"2026-04-04"},

{"title":"Short-prefix soft values: a null result","url":"/sections/science/2026/04/03/Short-prefix-soft-values-a-null-result.html","excerpt":"Abstract\n","categories":["sections","science","deep-learning"],"tags":[],"date":"2026-04-03"},

{"title":"Recursive decomposition with a continuation policy","url":"/sections/science/2026/04/03/Recursive-decomposition-with-continuation-policy.html","excerpt":"Abstract\n","categories":["sections","science","deep-learning"],"tags":[],"date":"2026-04-03"},

{"title":"MellowMax, Doob's h-transform, and the intensive geometry of tool-use agents","url":"/sections/science/2026/04/03/Mellowmax-doob-and-agentic-tool-use.html","excerpt":"Abstract\n","categories":["sections","science","deep-learning"],"tags":[],"date":"2026-04-03"},

{"title":"Soft values, symmetry breaking, and random rooted trees","url":"/sections/science/2026/04/02/Soft-values-symmetry-breaking-and-random-rooted-trees.html","excerpt":"A missing local law\n","categories":["sections","science","deep-learning"],"tags":[],"date":"2026-04-02"},

{"title":"Inference-Time Steering as a Discrete Schrödinger Bridge","url":"/sections/science/2026/04/01/Inference-time-steering-Schrodinger-Bridge.html","excerpt":"The limits of passive filtering\n","categories":["sections","science","deep-learning"],"tags":[],"date":"2026-04-01"},

{"title":"PLP is inference-time approximation of free energy","url":"/sections/science/2026/03/31/PLP-is-inference-time-approximation-of-free-energy.html","excerpt":"Approximate sampling and inference in LLMs\n","categories":["sections","science","deep-learning"],"tags":[],"date":"2026-03-31"},

{"title":"Test-time inference, language models and energy based models","url":"/sections/science/2026/03/27/PLP-and-Energy-Based-Models.html","excerpt":"The link between autoregressive models, energy based models and probabilistic language programming\n","categories":["sections","science","deep-learning"],"tags":[],"date":"2026-03-27"},

{"title":"Running Claude Code via OpenRouter in Docker","url":"/sections/tech/2026/03/16/Claude-code-inside-docker-container-with-openrouter.html","excerpt":"AI coding agents like Claude Code are incredibly powerful, but running them directly on your host machine can feel risky. \nContainerizing them is the logical step, but it often leads to auth conflicts, 401 errors, or terminal freezes.\n","categories":["sections","tech"],"tags":["docker","claude-code","openrouter","python"],"date":"2026-03-16"},

{"title":"Unleash Claude Code CLI with OpenRouter: Free AI Coding Power!","url":"/sections/tech/2026/03/15/Free-Claude-Code.html","excerpt":"Claude Code CLI with OpenRouter\n","categories":["sections","tech"],"tags":["claude-code","openrouter","cli"],"date":"2026-03-15"},

{"title":"Scaffolding Is All You Need","url":"/sections/science/2026/03/02/Scaffolding-is-all-you-need.html","excerpt":"+—\nlayout: post\ntitle: Scaffolding is all you need\ndescription: “When recursive LLM scaffolds improve reliability—and when they cannot.”\ndate: 2026-03-02\npublished: true\ncategories:\n\n  science\n  \n    deep-learning\n  \n\n","categories":["sections","science"],"tags":[],"date":"2026-03-02"},

{"title":"Probabilistic Language Programming, turning craft into solid science","url":"/sections/science/2026/03/01/Probabilistic-Language-Programming.html","excerpt":"Assigning semantic meaning to probabilistic programs\n","categories":["sections","science","deep-learning"],"tags":[],"date":"2026-03-01"},

{"title":"Intelligenza Artificial Neurosimbolica","url":"/sections/science/2025/12/26/Neurosymbolic-AI-at-rescue.html","excerpt":"Verso l’integrazione fra AI simbolica e AI neurale\n","categories":["sections","science","deep-learning"],"tags":[],"date":"2025-12-26"},

{"title":"Introducing skfolio new online convex optimization module","url":"/sections/science/2025/10/03/Introducing-skfolio-online-convex-optimization.html","excerpt":"Bridging online convex optimization into skfolio\n","categories":["sections","science","machine-learning"],"tags":[],"date":"2025-10-03"},

{"title":"Hard and soft analogies in machine learning","url":"/sections/science/2025/06/01/Hard-soft-analogies-machine-learning.html","excerpt":"In physics, the transition from a “hard” microcanonical ensemble to a “soft” canonical ensemble is governed by the introduction of temperature. In ML, we do the exact same thing to make functions differentiable, allowing gradients to flow.\n","categories":["sections","science","deep-learning"],"tags":[],"date":"2025-06-01"},

{"title":"Configuration-model Surprise for community detection","url":"/sections/science/2025/03/30/Configuration-model-surprise-for-community-detection.html","excerpt":"From vertex pairs to stubs\n","categories":["sections","science","complex-networks"],"tags":[],"date":"2025-03-30"},

{"title":"Running LM-Studio headlessly on Ubuntu 24.04 with CUDA","url":"/sections/tech/2025/03/12/Running-LMStudio-As-Local-Inference-Server.html","excerpt":"Running LM-Studio Headlessly on Ubuntu 24.04 with CUDA and Systemd\n","categories":["sections","tech"],"tags":[],"date":"2025-03-12"},

{"title":"Unlocking the Power of DeepSeek R1 via HuggingFace's TogetherAI","url":"/sections/tech/2025/01/30/Calling-deepseek-r1-from-cli.html","excerpt":"In the evolving landscape of artificial intelligence, the ability to interact with intelligent APIs has become a fundamental requirement for developers and organizations alike. Today, we explore how to harness the capabilities of the DeepSeek R1 model from HuggingFace through ...","categories":["sections","tech"],"tags":[],"date":"2025-01-30"},

{"title":"Prompt Chaining with Foundational models","url":"/sections/tech/2025/01/16/Prompt-chaining-with-foundational-models.html","excerpt":"In the rapidly evolving world of artificial intelligence and natural language processing, one of the most innovative concepts emerging is Prompt Chaining. This method optimizes the capabilities of Large Language Models (LLMs) by enabling a structured, step-by-step collaboratio...","categories":["sections","tech"],"tags":[],"date":"2025-01-16"},

{"title":"Creating the How2 Function for Natural Language Bash Commands with LLM","url":"/sections/tech/2025/01/15/How2-terminal-command.html","excerpt":"In today’s tech-driven world, the ability to translate natural language into executable commands is a powerful tool for developers and system administrators alike. Recently, I embarked on creating a function that bridges this gap using a Large Language Model (LLM) to interpret...","categories":["sections","tech"],"tags":[],"date":"2025-01-15"},

{"title":"Multilabel classification with Parabel and efficient at-k metrics","url":"/sections/science/2024/06/30/Multilabel-classification-with-Parabel-and-efficient-at-k-metrics.html","excerpt":"The annoying gap in extreme multilabel classification\n","categories":["sections","science","machine-learning"],"tags":[],"date":"2024-06-30"},

{"title":"PySpark initialization outside the Pyspark shell","url":"/sections/tech/2024/03/04/Initializing-Spark-with-findspark.html","excerpt":"How you do it with the findspark package without the need to startup a Spark shell with the options to load within a jupyterlab session.\n","categories":["sections","tech"],"tags":[],"date":"2024-03-04"},

{"title":"Installing XFormers on Mac M1/M2","url":"/sections/tech/2024/02/26/Installing-XFormers-on-Mac-M1.html","excerpt":"XFormers\n","categories":["sections","tech"],"tags":[],"date":"2024-02-26"},

{"title":"Calculate a file hash in Python","url":"/sections/tech/2023/01/14/Calculate-file-hash-in-python.html","excerpt":"This code helps you calculating the SHA256 hash of a file in Python. This could be helpful for the disambiguation of different files.\n","categories":["sections","tech"],"tags":[],"date":"2023-01-14"},

{"title":"Augmentation of documents for neural networks training","url":"/sections/tech/2023/01/14/Augmentation-of-documents-for-neural-networks-training.html","excerpt":"Why augmentation of documents\n","categories":["sections","tech"],"tags":[],"date":"2023-01-14"},

{"title":"Slicing rolling expanding windows over multiple pandas objects","url":"/sections/tech/2022/11/23/slicing-rolling-expanding-windows-over-multiple-pandas-objects.html","excerpt":"This is a fast way to yield a subset of rows from multiple Pandas dataframes or Series, when one needs to work on a sliding window basis over a predefined minimum and maximum number of rows. This approach is among the fastest available and is based on the .iloc accessor of bot...","categories":["sections","tech"],"tags":[],"date":"2022-11-23"},

{"title":"Dependencies needed for detectron2 on mac os","url":"/sections/tech/2022/09/15/Dependencies-for-having-detectron2-running-on-mac-os.html","excerpt":"Python dependencies for having detectron2 running on Mac OS\n","categories":["sections","tech"],"tags":[],"date":"2022-09-15"},

{"title":"How to enable colab for longer runtimes by keeping it active","url":"/sections/tech/2022/08/07/Enable-colab-for-longer-runtimes.html","excerpt":"This post is of help:\n","categories":["sections","tech"],"tags":[],"date":"2022-08-07"},

{"title":"How to download google sheet data to Pandas","url":"/sections/tech/2022/06/15/Downloading-Google-sheet-tables-to-pandas.html","excerpt":"All you need is a Google Sheets file with one or more sheets and of course some data. The file needs to be set to the sharing option which allows everyone with the link to view the data.\n","categories":["sections","tech"],"tags":[],"date":"2022-06-15"},

{"title":"Better graphics with matplotlib and seaborn","url":"/sections/tech/2022/05/03/Matplotlib-better-graphics.html","excerpt":"High–quality figures are not an aesthetic afterthought. In scientific communication they determine readability, reproducibility, and perceived rigor. Default matplotlib settings are optimized for quick inspection, not for publication. The configuration below standardizes typog...","categories":["sections","tech"],"tags":["matplotlib","helvetica"],"date":"2022-05-03"},

{"title":"Changepoint detection on huge grouped dataframes with ruptures and PySpark","url":"/sections/tech/2021/12/03/Scalable-changepoint-detection-on-huge-dataframes-with-ruptures-and-pyspark.html","excerpt":"Here we describe a way to perform scalable changepoint detection on grouped time series data by using PySpark and the rupture library.\n","categories":["sections","tech"],"tags":[],"date":"2021-12-03"},

{"title":"Rolling operations in PySpark","url":"/sections/tech/2021/12/03/Rolling-operations-in-PySpark.html","excerpt":"Rolling window operations are possible\n","categories":["sections","tech"],"tags":[],"date":"2021-12-03"},

{"title":"Sampling 10 rows per groupb in PySpark","url":"/sections/tech/2021/10/25/Sampling_10_rows_per_group_in_pyspark.html","excerpt":"You need to use a window partition by and let the random number do the shuffle for you.\n","categories":["sections","tech"],"tags":[],"date":"2021-10-25"},

{"title":"Docker on MacOS","url":"/sections/tech/2021/09/28/Docker-on-Mac-OS.html","excerpt":"How to install Docker on MacOs\n","categories":["sections","tech"],"tags":[],"date":"2021-09-28"},

{"title":"How to make HDBScan an inductive clustering method","url":"/sections/tech/2021/07/28/How-to-extend-hdbscan-to-make-it-an-inductive-clustering-method.html","excerpt":"There is a large difference between inductive and transductive clustering methods.\nWhile the first are more similar to supervised learning, in the sense that once trained on N examples they can generalize to M unseen new samples, transductive method instead need to see all dat...","categories":["sections","tech"],"tags":[],"date":"2021-07-28"},

{"title":"How to convert networkx graphs to graph-tool","url":"/sections/tech/2021/07/28/How-to-convert-networkx-graphs-to-graph-tool.html","excerpt":"How to convert networkx graphs to graph-tool\n","categories":["sections","tech"],"tags":[],"date":"2021-07-28"},

{"title":"Finding and deleting memory hungry temporary variables in jupyter notebooks","url":"/sections/tech/2021/06/21/Finding-and-deleting-memory-hungry-temporary-variables-in-jupyter-notebooks.html","excerpt":"This simple command displays the name of the variables in the current kernel in Jupyter notebooks, whichi are clogging our memory resources the most.\n","categories":["sections","tech"],"tags":[],"date":"2021-06-21"},

{"title":"Plotly and Jupyterlab issues","url":"/sections/tech/2021/05/18/Plotly-and-JupyterLab.html","excerpt":"How to install jupyterlab utilities\n","categories":["sections","tech"],"tags":[],"date":"2021-05-18"},

{"title":"The Woodbury matrix identity","url":"/sections/science/2020/10/23/The-Woodbury-matrix-identity.html","excerpt":"The Woodbury matrix identity is a useful identity in linear algebra.\nIt says that you can invert the sum of a matrix plus a \\(k\\)-rank correction by doing a rank \\(k\\)-correction to the inverse of the original matrix.\nIt is also called matrix inversion lemma or Sherman-Morriso...","categories":["sections","science","machine-learning"],"tags":[],"date":"2020-10-23"},

{"title":"Versioning data files for machine learning projects with DVC","url":"/sections/tech/2020/10/09/Versioning-data-files-for-machine-learning-projects.html","excerpt":"DVC Tutorial\n","categories":["sections","tech"],"tags":[],"date":"2020-10-09"},

{"title":"Prefixspan algorithm for frequent subsequences visualization","url":"/sections/science/2020/10/05/Prefixspan-algorithm-for-frequent-subsequences-visualization.html","excerpt":"Sequential pattern mining and projections\n","categories":["sections","science","machine-learning"],"tags":[],"date":"2020-10-05"},

{"title":"Code indentation in Python with Black","url":"/sections/tech/2020/09/04/Code-indentation-in-Python-with-black.html","excerpt":"A very nice code indentation tool for Python is called Black\n","categories":["sections","tech"],"tags":[],"date":"2020-09-04"},

{"title":"How to merge Dataframes in spark as in Pandas","url":"/sections/tech/2020/05/18/merge_dataframes_in_spark_as_in_pandas.html","excerpt":"This is how you do it:\n","categories":["sections","tech"],"tags":[],"date":"2020-05-18"},

{"title":"Some scala functions as exercise","url":"/sections/tech/2020/04/20/Some-scala-functions-as-exercise.html","excerpt":"Here are some very basic function with self-explanatory name, to be coded in Scala for a super beginner.\nSome of them are very inefficient, pay attention to use them, they are only for demonstration.\n","categories":["sections","tech"],"tags":[],"date":"2020-04-20"},

{"title":"Installing Jupyter with a Scala + Spark kernel","url":"/sections/tech/2020/04/01/Installing-Jupyter-with-scala-kernel.html","excerpt":"Installazione SCALA + SPARK + Jupyter\n","categories":["sections","tech"],"tags":[],"date":"2020-04-01"},

{"title":"How to remove all logging information in scala spark","url":"/sections/tech/2020/03/27/How-to-remove-all-logging-information-in-spark-scala.html","excerpt":"Create a folder named “log4j” in the root folder of your project (the one where build.sbt stays) and then create a file named “log4j.properties” with the following content:\n","categories":["sections","tech"],"tags":[],"date":"2020-03-27"},

{"title":"Eigenvectors from eigenvalues a numpy implementation","url":"/sections/science/2019/11/14/Eigenvectors-from-eigenvalues-A-numpy-implementation.html","excerpt":"Simple ideas can make good numerical algorithms\n","categories":["sections","science","machine-learning"],"tags":[],"date":"2019-11-14"},

{"title":"Sampling uniform spanning forests with Wilson algorithm in Python","url":"/sections/science/2019/09/03/Sampling-uniform-random-spanning-forests-with-Wilson-algorithm.html","excerpt":"Stochastic ways to determine the laplacian spectrum\n","categories":["sections","science","complex-networks"],"tags":[],"date":"2019-09-03"},

{"title":"The link between machine learning and statistical physics","url":"/sections/science/2019/06/12/The-link-between-machine-learning-and-statistical-physics.html","excerpt":"Reading the paper by Max Tegmark “Why does deep and cheap learning work so well” is illuminating.\n","categories":["sections","science","deep-learning"],"tags":[],"date":"2019-06-12"},

{"title":"How to expand the loopback device size","url":"/sections/tech/2019/05/03/increasing-loopback-device-space-when-used-as-dropbox-ext4-filesystem.html","excerpt":"From this answer\n","categories":["sections","tech"],"tags":[],"date":"2019-05-03"},

{"title":"Quantum statistical mechanics of complex networks in networkqit","url":"/sections/science/2019/04/18/Introducing-networkqit.html","excerpt":"With this post I am introducing to the public my very first complete Python package networkqit.\n","categories":["sections","science","complex-networks"],"tags":[],"date":"2019-04-18"},

{"title":"Hidden Markov Models or HMM","url":"/sections/science/2019/03/21/Hidden-markov-models.html","excerpt":"Hidden Markov Models are at the core of many machine learning techniques\n","categories":["sections","science","machine-learning"],"tags":[],"date":"2019-03-21"},

{"title":"Short introduction to measure theory","url":"/sections/science/2019/03/18/Basic-properties-of-measures.html","excerpt":"In these notes, we’ll take a look at the bare essentials of modern probability theory.\nWe’ll describe the basic ideas, but we will skip Lebesgue integration, at the moment.\nTo do this, we need to introduce some measure theory first, to see how things emerge naturally and beati...","categories":["sections","science","machine-learning"],"tags":[],"date":"2019-03-18"},

{"title":"Low rank estimation of sparse observed ratios","url":"/sections/science/2019/03/13/Rank-1-matrix-estimation.html","excerpt":"Reccomender systems are a class of algorithms to deal with missing information.\nGiven that we have some available rate about the relations of a set of objects, and these informations are specified by real numbers, how can we estimate the relations between another subset of obj...","categories":["sections","science","machine-learning"],"tags":[],"date":"2019-03-13"},

{"title":"Exercises for quantitative interviews","url":"/sections/finance/2019/02/04/Exercises-for-quant-interviews.html","excerpt":"Exercise 1\n\n","categories":["sections","finance"],"tags":[],"date":"2019-02-04"},

{"title":"Smooth approximation to the floor function","url":"/sections/science/2019/01/25/Smooth-approximation-to-the-floor-function.html","excerpt":"I need to sample random numbers distributed according to the geometric distribution.\nSimilarly to the Box-Muller transformation, which is a method to sample normally distributed random numbers based on a uniform random generator,\nI have found that any probability distribution ...","categories":["sections","science","machine-learning"],"tags":[],"date":"2019-01-25"},

{"title":"Visual inspection of python profiler output with gprof2dot","url":"/sections/tech/2019/01/17/visual-inspection-of-python-profiling-output.html","excerpt":"Ever wondered how to make a beatiful and powerful call graph from your python profiler?\nI have found that there is no need to install heavyweight tools like KCachegrind or even worse, buying expensive IDEs.\nThe solution is simple.\n","categories":["sections","tech"],"tags":[],"date":"2019-01-17"},

{"title":"Einstein summation in Numpy","url":"/sections/science/2019/01/16/Einstein-summation-in-numpy-and-tensorflow.html","excerpt":"Einstein summation is a convention in tensor algebra where repeated indices are implicitly summed.\nFor example, imagine we have a matrix (a tensor of rank 2) \\(A_{i,j}\\). To compute the trace, i.e. the sum of diagonal elements one has to compute\n","categories":["sections","science","machine-learning"],"tags":[],"date":"2019-01-16"},

{"title":"Showing brain parcellation in Python with nilearn and some hacking","url":"/sections/science/2018/11/27/Plotting-custom-brain-parcellation-beautifully-in-python-with-nilearn.html","excerpt":"In this post I would like to introduce to the nilearn user, a modified set of functions based on the nilearn.surface module, that are of great help in making beautiful surface colored pictures of brain, like the one in this figure:\n","categories":["sections","science","machine-learning"],"tags":[],"date":"2018-11-27"},

{"title":"Faceting images or generic plots with Seaborn and Python matplotlib","url":"/sections/tech/2018/11/12/Faceting-images-with-matplotlib-and-seaborn.html","excerpt":"I’ve found myself working with large pandas dataframe.\nDifferently from the typical usage of pandas dataframes, in some cells I have numpy.array as content, or other types of data.\nHere we call these non-standard columns as x and y.\n","categories":["sections","tech"],"tags":[],"date":"2018-11-12"},

{"title":"Install latest igraph 0.7.1 for Python3 on Ubuntu","url":"/sections/tech/2018/10/23/Install-igraph-for-python3-on-ubuntu.html","excerpt":"If you choose to use the igraph library with Python 2, it’s a cakewalk to get it running on a fresh install of Ubuntu 16.04:\n","categories":["sections","tech"],"tags":[],"date":"2018-10-23"},

{"title":"Weighted graphs from adjacency matrix in graph-tool","url":"/sections/science/2018/09/12/weighted-graph-from-adjacency-matrix-in-graph-tool.html","excerpt":"I was playing a bit with networks in Python. In my daily life I typically work with adjacency matrices, rather than other sparse formats for networks.\nAdjacency matrix is pretty good for visualization of communities, as well as to give an idea of the distribution of edge weigh...","categories":["sections","science","complex-networks"],"tags":[],"date":"2018-09-12"},

{"title":"How to list all tensorflow devices","url":"/sections/tech/2018/07/10/How-to-find-all-tensorflow-devices.html","excerpt":"This is how you do:\n","categories":["sections","tech"],"tags":[],"date":"2018-07-10"},

{"title":"Random matrix ensemble spectral density and the average resolvent","url":"/sections/science/2018/06/25/spectral-density-via-average-resolvent.html","excerpt":"This short blog note is covering some aspects related to interesting calculations that can be done in random matrix theory applied to the study of spectral properties of random graph models, like those shown in Figure:\n","categories":["sections","science","complex-networks"],"tags":[],"date":"2018-06-25"},

{"title":"Some considerations about the spectral entropies framework","url":"/sections/science/2018/06/24/some-considerations-about-spectral-entropies-framework.html","excerpt":"I am illustrating some of the ideas that emerged before, during and after the conference.\nHow to compute this quantity?\n","categories":["sections","science","complex-networks"],"tags":[],"date":"2018-06-24"},

{"title":"Dataset of biologically derived graphs","url":"/sections/science/2018/05/21/Dataset-of-biologically-derived-graphs.html","excerpt":"This website contains a number of interesting dataset of biologically derived graphs.\n","categories":["sections","science","complex-networks"],"tags":[],"date":"2018-05-21"},

{"title":"Why probabilistic programming matters","url":"/sections/science/2018/05/21/why-probabilistic-programming-matters.html","excerpt":"I recently stumbled across the idea of probabilistic programming, something I had never heard before.\nIndeed with the recent advances in artificial intelligence and machine learning, this field is seeing a very rapid development.\nThe motivation why I found this research field ...","categories":["sections","science","deep-learning"],"tags":[],"date":"2018-05-21"},

{"title":"Inequalities, traces, densities, matrices","url":"/sections/science/2018/05/11/Inequalities-traces-densities-matrices.html","excerpt":"Let us compute the following quantity:\n","categories":["sections","science","machine-learning"],"tags":[],"date":"2018-05-11"},

{"title":"An identity for zero","url":"/sections/science/2018/05/11/An-identity-for-zero.html","excerpt":"This identity could result useful\n","categories":["sections","science","machine-learning"],"tags":[],"date":"2018-05-11"},

{"title":"Distance matrix of the air500 complex network","url":"/sections/science/2018/03/12/Distance-matrix-of-the-air500-complex-network.html","excerpt":"The air500 network is the adjacency matrix of the largest 500 airports in the world.\n","categories":["sections","science","complex-networks"],"tags":[],"date":"2018-03-12"},

{"title":"Lambda functions composition in Python 3","url":"/sections/science/2018/03/01/Lambda-functions-composition-in-python3.html","excerpt":"Introduction\n","categories":["sections","science","machine-learning"],"tags":[],"date":"2018-03-01"},

{"title":"A Graduate Course in Econometrics","url":"/sections/finance/2017/09/27/A-Graduate-Course-in-Econometrics.html","excerpt":"Introduction to matrix econometrics\n","categories":["sections","finance"],"tags":[],"date":"2017-09-27"},

{"title":"A Full Undergraduate Course in Econometrics","url":"/sections/finance/2017/09/22/A-Full-Undegraduate-Course-In-Econometrics.html","excerpt":"This series of notes is from the Ben Lambert’s course on econometrics, a discipline mixing calculus, linear algebra and statistics in a way that I believe is very enjoyable for a physicist-minded person.\nThe series of courses is available at the following Youtube link:\n","categories":["sections","finance"],"tags":[],"date":"2017-09-22"},

{"title":"The Enhanced weighted random graph model (EWRG)","url":"/sections/science/2017/05/12/Enhanced-weighted-random-graph-model-(EWRG).html","excerpt":"The ensemble of maximally random weighted graphs with the constant average number of edges and total weight.\n","categories":["sections","science","complex-networks"],"tags":[],"date":"2017-05-12"},

{"title":"Plugging the weighted random graph into Surprise","url":"/sections/science/2017/05/10/Surprise-to-support-weighted-random-graph.html","excerpt":"Surprise is based on the calculation of the number of simple graphs with \\(n\\) nodes and \\(m\\) edges exactly.\nThis null model is called \\(G_{nm}\\) model and it is the microcanonical version of the Erdos-Renyi model also called \\(G_{np}\\).\nTo be more precise, Surprise does not ...","categories":["sections","science","complex-networks"],"tags":[],"date":"2017-05-10"},

{"title":"Statistical mechanics of networks - Park and Newman model of hidden variables","url":"/sections/science/2017/05/09/Statistical-mechanics-of-networks.html","excerpt":"Introduction\n","categories":["sections","science","complex-networks"],"tags":[],"date":"2017-05-09"},

{"title":"How to stop with CTRL C a running mex file in Matlab","url":"/sections/tech/2017/02/12/How-to-stop-with-ctrlc-a-running-mex-file-in-matlab.html","excerpt":"This is how you do:\n","categories":["sections","tech"],"tags":[],"date":"2017-02-12"},

{"title":"Eigenvalue spectra of modular matrix techniques and calculations","url":"/sections/science/2017/02/12/Eigenvalue-spectra-of-modular-matrix-techniques.html","excerpt":"RANDOM NOTES ABOUT RANDOM GRAPHS AND EIGENVALUES, NOT TO BE TAKEN SERIOUSLY\n","categories":["sections","science","complex-networks"],"tags":[],"date":"2017-02-12"},

{"title":"How to grant specific directory access to users in linux","url":"/sections/tech/2017/01/27/How-to-grant-specific-directory-access-to-users-in-linux.html","excerpt":"Negative ACLs\nYou can prevent a user from accessing certain parts of the filesystem by setting access control lists. For example, to ensure that the user abcd cannot access any file under /home:\n","categories":["sections","tech"],"tags":[],"date":"2017-01-27"},

{"title":"Assigning XRDP to different users","url":"/sections/tech/2017/01/20/Assigning-XRDP-to-different-users.html","excerpt":"/etc/rc.local \nwrite\nxrdp-sesrun 127.0.0.1 brainet neuron128 1280 1024 24\n","categories":["sections","tech"],"tags":[],"date":"2017-01-20"},

{"title":"How to create a large swap file in Ubuntu if system runs out of memory.","url":"/sections/tech/2017/01/17/Create-a-large-swap-file-in-ubuntu.html","excerpt":"Check this guide:\n","categories":["sections","tech"],"tags":[],"date":"2017-01-17"},

{"title":"TAB is not working with XFCE4 after installation from Ubuntu for use with remote desktop","url":"/sections/tech/2017/01/17/TAB-not-working-in-xfce4-with-remote-desktop-after-ubuntu-installation.html","excerpt":"You have installed Ubuntu 14.04 or latest Ubuntu LTS 16.04 on your server and you want to make it available for remote desktop connections with the help of XRDP.\nYou install xrdp and you connect to your server with some client, like remmina.\nSuddenly XFCE4 is showing up in its...","categories":["sections","tech"],"tags":[],"date":"2017-01-17"},

{"title":"Matlab and CUDA gpudevice freezes on Ubuntu 14.04 with GTX 1070","url":"/sections/tech/2017/01/16/Matlab-and-CUDA-gpudevice-freezes-on-Ubuntu-with-GTX1070.html","excerpt":"In Ubuntu 14.04 with MATLAB R2016b it may happen that after an installation of CUDA8 and NVidia drivers 367, the calls to specific CUDA functions such as gpudevice or gpuArray are very slow the first time you call them.\nThis is because MATLAB R2016b it’s not built against the ...","categories":["sections","tech"],"tags":[],"date":"2017-01-16"},

{"title":"Computing Euler angles from 3x3 rotation matrix in Matlab","url":"/sections/science/2016/12/24/Computing-Euler-Angles.html","excerpt":"Computing Euler angles from a rotation matrix is straightforward once you set a convention. Indeed is possible to compute an entire different set of angles that defines a rotation when you change axis. In this case I use the aeronautical notation, with pitch, yaw and roll as a...","categories":["sections","science","machine-learning"],"tags":[],"date":"2016-12-24"},

{"title":"Computing hypergeometric probability efficiently in C++","url":"/sections/science/2016/10/28/Computing_hypergeometric_probability_efficiently_in_C++.html","excerpt":"Computing hypergeometric function is a slow and difficult process, often affected by overflow errors as evaluating binomial coefficient may return extremely large numbers.\nFortunately, thanks to some hypergeometric identities, is possible to evaluate the hypergeometric probabi...","categories":["sections","science","machine-learning"],"tags":[],"date":"2016-10-28"},

{"title":"How to start a SSH daemon on Windows without agonizing pain","url":"/sections/tech/2016/10/06/How-to-start-a-SSH-daemon-on-Windows-without-agonizing-pain.html","excerpt":"I recently needed to connect via ssh to my windows computer from a Linux server, but I couldn’t figure how to do it since ssh servers for windows are expensive and I don’t want to pay for a software that on Linux comes for free. Some of the typical ssh servers (daemon in linux...","categories":["sections","tech"],"tags":[],"date":"2016-10-06"},

{"title":"How to upgrade R to the latest version on Ubuntu 14.04","url":"/sections/tech/2016/10/06/How-to-upgrade-R-to-the-latest-version-on-Ubuntu-14.04.html","excerpt":"Follow this instruction:\n","categories":["sections","tech"],"tags":[],"date":"2016-10-06"},

{"title":"A code for the absolute orientation problem with Umeyama algorithm in Python","url":"/sections/tech/2016/10/06/A-code-for-absolute-orientation-problem-with-Umeyama-algorithm-in-Python.html","excerpt":"A code for the absolute orientation problem solved with Umeyama algorithm.\nExplanations, briefly.\n","categories":["sections","tech"],"tags":[],"date":"2016-10-06"},

{"title":"Delaunay triangulation and beautiful visual effects in Latex","url":"/sections/tech/2016/09/13/Delaunay-triangulation-and-beautiful-visual-effects-in-Latex.html","excerpt":"Recently in the world of the Internet this very fancy coloured pattern appeared:\n","categories":["sections","tech"],"tags":[],"date":"2016-09-13"},

{"title":"Comparing spectral densities of random graph models.","url":"/sections/science/2016/08/26/Spectral_density_random_graph_models_planted_partition.html","excerpt":"The ability of quantitatively comparing two graphs is of great importance in many scientific questions and of exceptional importance in studying brain networks.\nHow the brain networks of healthy people differ from those of patients? What kind of alterations are present in the ...","categories":["sections","science","complex-networks"],"tags":[],"date":"2016-08-26"},

{"title":"Matlab R2016a crashes on Ubuntu 16.04 with NVidia 361 drivers","url":"/sections/tech/2016/08/17/Matlab-R2016a-crashes-on-Ubuntu-16.04-with-NVidia-361-drivers.html","excerpt":"For those of you who upgraded Ubuntu from 14.04 to 16.04 and have found that Matlab is crashing with errors in nvidia drivers, you have three options:\n","categories":["sections","tech"],"tags":[],"date":"2016-08-17"},

{"title":"Fix black screen after upgrade from Ubuntu 14.04 to Ubuntu 16.04","url":"/sections/tech/2016/08/10/Fix-black-screen-after-upgrade-from-ubuntu-14-04-to-ubuntu-16-04.html","excerpt":"I’ve just decided to upgrade my Ubuntu 14.04.4 to the latest Ubuntu distribution 16.04.1. To do this I’ve followed the instructions and after the installation  rebooted my computer.\n","categories":["sections","tech"],"tags":[],"date":"2016-08-10"},

{"title":"Using-VNC-persistent-sessions","url":"/sections/tech/2016/08/02/Using-VNC-persistent-sessions.html","excerpt":"In this guide I briefly explain how you can get remote persistent session on the Mattarello Linux servers from any local computer, Windows, Linux or OSX.\n","categories":["sections","tech"],"tags":[],"date":"2016-08-02"},

{"title":"Setting up VNC sessions in Linux","url":"/sections/tech/2016/08/01/Setting-up-VNC-sessions-in-Linux.html","excerpt":"Server side\n","categories":["sections","tech"],"tags":[],"date":"2016-08-01"},

{"title":"Launching-matlab-in-background-correctly-in-linux","url":"/sections/tech/2016/07/16/Launching-matlab-in-background-correctly-in-linux.html","excerpt":"You want to launch Matlab for a very long script and then disconnect your remote terminal, and when back, not finding bad surprise.\n","categories":["sections","tech"],"tags":[],"date":"2016-07-16"},

{"title":"How to install latest R on servers under proxy","url":"/sections/tech/2016/06/27/How-to-install-latest-R-on-servers-under-proxy.html","excerpt":"It’s possible to keep your R version updated on Ubuntu 14.04 if you decide not to use the version provided by the package maintainer. In this small guide I explain how to do that especially if you are under a proxy server.\n","categories":["sections","tech"],"tags":[],"date":"2016-06-27"},

{"title":"Get the exact print size of a PDF in inches","url":"/sections/tech/2016/06/16/Get-Exact-Size-pdf-inches.html","excerpt":"This is the command to do the stuff\n","categories":["sections","tech"],"tags":[],"date":"2016-06-16"},

{"title":"Forwarding-X11-maintaining-connection-with-xpra","url":"/sections/tech/2016/06/16/Forwarding-X11-maintaining-connection-with-xpra.html","excerpt":"How to use XPRA for remote connection to the NeuralComputation Linux servers hosted in Mattarello\n","categories":["sections","tech"],"tags":[],"date":"2016-06-16"},

{"title":"Install Google Drive for Linux","url":"/sections/tech/2015/10/14/Install-Google-Drive-On-Linux.html","excerpt":"Gsync is the rsync for Google Drive. If you like me, have unlimited storage space on Google Drive, this guide that can be very useful to you. You can store your precious date on the google cloud with the warranty that they are not lost nor destroyed. Google treats data very se...","categories":["sections","tech"],"tags":[],"date":"2015-10-14"},

{"title":"Detecting the consensus communities in graphs","url":"/sections/science/2015/10/14/Detecting-the-consensus-communities-in-graphs.html","excerpt":"In these section we’ll address in depth an approach to making sense of\nthe mesoscopic structure of a network by means of non-deterministic\nmethods. We will take advantage of the methods of statistical physics\nand treat with ensembles of partitions to assess the statistical\nsig...","categories":["sections","science","complex-networks"],"tags":[],"date":"2015-10-14"},

{"title":"How to install Octave on OSX Yosemite","url":"/sections/tech/2015/10/13/How-to-install-Octave-on-OSX-Yosemite.html","excerpt":"For those who have problems installing Octave on OSX Yosemite, this is a simple guide.\n","categories":["sections","tech"],"tags":[],"date":"2015-10-13"},

{"title":"How to reset unity in Ubuntu 14.04","url":"/sections/tech/2015/10/05/How-to-reset-unity-in-Ubuntu-14.04.html","excerpt":"For those who struggle with frequent freezes of Unity, this is a short guide on how to reset Unity to default settings in Ubuntu 14.04.\n","categories":["sections","tech"],"tags":[],"date":"2015-10-05"},

{"title":"Matlab imagesc with text values","url":"/sections/tech/2015/09/25/imagesc-with-text-values.html","excerpt":"Well. Finally got around to making a better imagesc function in Matlab and Octave. I’ve named it imagesctxt and it has the same arguments as imagesc.\n","categories":["sections","tech"],"tags":[],"date":"2015-09-25"}

]