networks on Dan MacKinlay
https://danmackinlay.name/tags/networks.html
Recent content in networks on Dan MacKinlayHugo -- gohugo.ioen-usTue, 30 Mar 2021 11:19:44 +1100Science; Sociology and institution design for
https://danmackinlay.name/notebook/peer_review.html
Tue, 30 Mar 2021 11:19:44 +1100https://danmackinlay.name/notebook/peer_review.htmlOpen review processes, practical Mathematical models of the reviewing process Economics of publishing Mechanism desoign for peer review process How well does academia gatekeep? Style guide for reviews and rebuttals References Upon the thing that I presume academic publishing is supposed to do: further science. Reputation system and other mechanisms for trust in science, a.k.a. collective knowledge for reality itself.
I would like to consider the system of peer review, networking, conferencing, publishing and acclaim and see how closely it approximates an ideal system for uncovering truth, and further, imagine how we could make a better system.Filter bubbles, fact checking and kompromat
https://danmackinlay.name/notebook/filter_bubbles.html
Sun, 21 Mar 2021 15:51:21 +1100https://danmackinlay.name/notebook/filter_bubbles.htmlReferences News media and public shared reality. Fake news, incomplete news, alternative facts, strategic inference, kompromat, agnotology, facebooking to a molecular level. Basic media literacy and whether it helps. As seen in elections, and provocateur twitter bots.
Theirtube
Theirtube is a Youtube filter bubble simulator that provides a look into how videos are recommended on other people’s YouTube. Users can experience how the YouTube home page would look for six different personas.Diagramming and visualising graphical models
https://danmackinlay.name/notebook/diagrams_graphical_models.html
Mon, 15 Mar 2021 16:44:16 +1100https://danmackinlay.name/notebook/diagrams_graphical_models.htmlDaggity dagR yEd diagrammeR flowchart.fun Mermaid TETRAD Matplotlib Graphviz tikz Misc References On the art and science of algorithmic line drawings for representing graphical models, which is a important part of statistics. The diagrams we need here are nearly flowchart-like, so I can sketch them with a flowchart if need be; but they are closely integrated with the equations of a particular statistical model, so I would like to incorporate them into the same system to avoid tedious and error-prone manual sync.Learning on manifolds
https://danmackinlay.name/notebook/learning_on_manifolds.html
Wed, 03 Mar 2021 12:29:39 +1100https://danmackinlay.name/notebook/learning_on_manifolds.htmlLearning on a given manifold Information Geometry Hamiltonian Monte Carlo Langevin Monte Carlo Natural gradient Homogeneous probability References Abraham Bosse, Moyen vniuersel de pratiquer la perspectiue sur les tableaux, ou surfaces irregulieres : ensemble quelques particularitez concernant cet art, & celuy de la graueure en taille-douce (1653)
A placeholder for learning on curved spaces. Not discussed: learning OF the curvature of spaces.
Learning on a given manifold Learning where there is an a priori manifold seems to also be a usage here?Red queen social signal dynamics
https://danmackinlay.name/notebook/red_queen_signalling.html
Sat, 27 Feb 2021 11:11:16 +1100https://danmackinlay.name/notebook/red_queen_signalling.htmlNote salad Meme dynamics Martial arts In consumer design References Signalling games, subcultures, slang, El Farol Bars, anti-inductive systems, glass bead games, fake martial arts, level of simulacra, genre speciation, as transmitted on the social information graph …
Note salad Robin Hanson argues against irony for being outgroup-exclusionary. I don’t think blanket discouraging irony is plausible or desirable, but… the insight is useful. It is important to remember that indicators of in-group membership, such as irony, are shibboleths, not indicators of quality.Memetics
https://danmackinlay.name/notebook/memetics.html
Sat, 27 Feb 2021 11:11:07 +1100https://danmackinlay.name/notebook/memetics.htmlIncoming links Girardian mimetic violence Pluralistic ignorance Toxoplasma of rage References Placeholder.
Clara Vandeweerdt
I work on public opinion, media and climate politics. My latest paper talks about how reporting on events looks very different depending on the ideological color of the media outlet.
Incoming links YY Ahn
Giulio Rossetti, author of various network analysis libraries such as dynetx and ndlib.
Francesco BonchiSocial norms
https://danmackinlay.name/notebook/social_norms.html
Thu, 25 Feb 2021 17:28:51 +1100https://danmackinlay.name/notebook/social_norms.htmlReferences Asch conformity experiments, Schelling points, illusory norms via pluralistic ignorance or majority illusions.
Anders Sandberg, on AI versus social norm enforcemant:
We are subject to norm enforcement from friends and strangers all the time. What is new is the application of media and automation. They scale up the stakes and add the possibility of automated enforcement […] . Automated enforcement makes the panopticon effect far stronger: instead of suspecting a possibility of being observed it is a near certainty.Journalism, normative
https://danmackinlay.name/notebook/renewing_journalism.html
Fri, 19 Feb 2021 08:26:58 +1100https://danmackinlay.name/notebook/renewing_journalism.htmlReferences I do not have time for this now and so will say little apart from bookmarking items for future return. I would like to have a better notion of what systems of incentives we might submit ourselves to in order to gbe able to make a better claim that our journalism is informing us about the world in the manner in which we need it.Causal inference in the continuous limit
https://danmackinlay.name/notebook/causality_continuous.html
Wed, 17 Feb 2021 20:41:08 +1100https://danmackinlay.name/notebook/causality_continuous.htmlReferences Causality on continuous index spaces, and, which turns out to be related, equilibrium/feedback dynamics. Placeholder.
Bongers and Mooij (2018):
Uncertainty and random fluctuations are a very common feature of real dynamical systems. For example, most physical, financial, biochemical and engineering systems are subjected to time-varying external or internal random disturbances. These complex disturbances and their associated responses are most naturally described in terms of stochastic processes.Weaponised social media
https://danmackinlay.name/notebook/weaponized_social_media.html
Tue, 01 Dec 2020 08:18:00 +1100https://danmackinlay.name/notebook/weaponized_social_media.htmlConspiracy theories and their uses Partisanship in social media firms Rhetorical strategies Evaluating effectiveness Automatic trolling, infinite fake news Post hoc analysis Incoming think pieces References Information warfare on the social information graph, for the purpose of human behaviour control, various notes to that theme.
The other side to trusted news; hacking the implicit reputation system of social media to suborn factual reporting, or to motivate people to behave to suit your goals, to, e.Recommender systems
https://danmackinlay.name/notebook/recommender_systems.html
Mon, 30 Nov 2020 14:55:18 +1100https://danmackinlay.name/notebook/recommender_systems.htmlReferences Not my area, but I need a landing page to refer to for some non-specialist contacts of mine.
I am most familiar with the matrix factorization approaches (e.g. factorization machines, NNMF) but there are many, e.g. variational autoencoder approaches are en vogue.
An overview by Javier lists many approaches.
Most Popular recommendations (the baseline) Item-User similarity based recommendations kNN Collaborative Filtering recommendations GBM based recommendations Non-Negative Matrix Factorization recommendations Factorization Machines (Steffen Rendle 2010) Field Aware Factorization Machines (Yuchin Juan, et al, 2016) Deep Learning based recommendations (Wide and Deep, Heng-Tze Cheng, et al, 2016) Neural Collaborative Filtering (Xiangnan He et al.Variational inference by message-passing in graphical models
https://danmackinlay.name/notebook/message_passing.html
Wed, 25 Nov 2020 17:42:32 +1100https://danmackinlay.name/notebook/message_passing.htmlReferences Variational inference where the model factorizes over some graphical independence structure, which means we get cheap and distributed inference. I am currently particularly interested in this for latent GP models. Many things can be expressed as message passing algorithms. The grandparent idea in this unification seems to be “Belief propagation”, a.k.a. “sum-product message-passing”, credited to (Pearl, 1982) for DAGs and then generalised to MRFs, PGMs, factor graphs etc.Graph neural nets
https://danmackinlay.name/notebook/nn_graph.html
Tue, 24 Nov 2020 08:33:48 +1100https://danmackinlay.name/notebook/nn_graph.htmlReferences Neural networks applied to graph data. (Neural networks of course can already be represented as directed graphs, or applied to phenomena which arise from a causal graph but that is not what we mean here.
The version of graphical neural nets with which I am familiar is applying convnets to spectral graph representations. e.g. Thomas Kipf summarises research there.
I gather that the field has moved on and I am no longer across what is happening.External validity
https://danmackinlay.name/notebook/external_validity.html
Mon, 09 Nov 2020 15:58:56 +1100https://danmackinlay.name/notebook/external_validity.htmlStandard graphical models Tools Salad Meta References TBD.
This Maori gentleman from the 1800s demonstrates an artful transfer learning from the western fashion domain
One could read Sebastian Ruder’s NN-style introduction to “transfer learning”. NN people like to think about this in particular way which I like because of the diversity of out-of-the-box ideas it invites and which I dislike because it is sloppy.Causal inference on DAGs
https://danmackinlay.name/notebook/causal_inference.html
Wed, 04 Nov 2020 12:36:13 +1100https://danmackinlay.name/notebook/causal_inference.htmlLearning materials do-calculus Counterfactuals Continuously indexed fields External validity Propensity scores Causal Graph inference from data Causal time series DAGS Drawing graphical models Tools References Inferring the optimal intervention requires accounting for which arrows are independent of which
Inferring cause and effect from nature. Graphical models and related techniques for doing it. Avoiding the danger of folk statistics. Observational studies, confounding, adjustment criteria, d-separation, identifiability, interventions, moral equivalence…Efficient factoring of GP likelihoods
https://danmackinlay.name/notebook/gp_factoring.html
Mon, 26 Oct 2020 12:46:34 +1100https://danmackinlay.name/notebook/gp_factoring.htmlBasic sparsity via inducing variables SVI for Gaussian processes Latent Gaussian Process models References There are many ways to cleverly slice up GP likelihoods so that inference is cheap.
This page is about some of them, especially the union of sparse and variational tricks. Scalable Gaussian process regressions choose cunning factorisations such that the model collapses down to a lower-dimensional thing than it might have seemed to need, at least approximately.The levels of simulacra
https://danmackinlay.name/notebook/simulacra.html
Mon, 19 Oct 2020 07:46:25 +1100https://danmackinlay.name/notebook/simulacra.htmlZvi Mowshowitz has some interesting models of how truth-telling works, but because he is, er, loquacious, it helps to have shorter versions of his posts to refer to. Which I might do here.
For now, though, I’ll simply link to the levels of simulacra model. In the case of modelling COVID there are some interesting analyses of truth dynamics hypothesised:
To the purely political actor, the implausible lie is better.Graph computation
https://danmackinlay.name/notebook/graph_computation.html
Sun, 20 Sep 2020 07:49:00 +1000https://danmackinlay.name/notebook/graph_computation.htmlReferences Engines for calculating things on or about graphs. Which is everything, in a trivial case, but usually when we talk about graph computation we mean things that are more simply or more elegantly represented as graphs, which usually implies having some kind of sparsity in edges.
For specific applications of such computations, see e.g. graphical models or complex networks, inference on social graphs etc. Sometimes we use a linear algebra representation of a graph connectivity pattern, e.Causal inference in highly parameterized ML
https://danmackinlay.name/notebook/causality_ml.html
Fri, 18 Sep 2020 09:34:46 +1000https://danmackinlay.name/notebook/causality_ml.htmlReferences TBD.
Léon Bottou, From Causal Graphs to Causal Invariance
For many problems, it’s difficult to even attempt drawing a causal graph. While structural causal models provide a complete framework for causal inference, it is often hard to encode known physical laws (such as Newton’s gravitation, or the ideal gas law) as causal graphs. In familiar machine learning territory, how does one model the causal relationships between individual pixels and a target prediction?Applied string mangling
https://danmackinlay.name/notebook/string_mangling.html
Mon, 14 Sep 2020 17:12:39 +1000https://danmackinlay.name/notebook/string_mangling.htmlRegexp Parsers A.k.a. Un-natural language processing.
Regexp Image used under CC licence from Martin Haverbeke’s Eloquent Javascript.
A.k.a. regexes. A.k.a. “regular expressions”, from a principled origin they presumably had in the theory of syntax. However, regexes as commonly encountered encode a particular way of specifying a language, rather than some arbitrary class of regular languages.
The default flavour of string matching, available in a variety of flavours, all equally boring.Independence, conditional, statistical
https://danmackinlay.name/notebook/independence.html
Sun, 13 Sep 2020 08:43:52 +1000https://danmackinlay.name/notebook/independence.htmlAs an algebra Tests Traditional tests Chatterjee ξ Copula tests Information criteria Kernel distribution embedding tests Stein Discrepancy References Conditional independence between random variables is a special relationship. As seen in inference directed graphical models.
Connection with model selection, in the sense that accepting enough true hypotheses leaves you with a residual independent of the predictors. (🏗 clarify.)
David Butler on independence venn diagramsDimensionality reduction
https://danmackinlay.name/notebook/dimensionality_reduction.html
Fri, 11 Sep 2020 08:20:03 +1000https://danmackinlay.name/notebook/dimensionality_reduction.htmlBayes Learning a summary statistic Feature selection PCA and cousins Learning a distance metric UMAP For indexing my database Locality Preserving projections Diffusion maps As manifold learning Multidimensional scaling Random projection Stochastic neighbour embedding and other visualisation-oriented methods Autoencoder and word2vec Misc References 🏗🏗🏗🏗🏗
I will restructure learning on manifolds and dimensionality reduction into a more useful distinction.
You have lots of predictors in your regression model!Causal graphical model reading group 2020
https://danmackinlay.name/post/reading_group_2020_causal_dags.html
Thu, 03 Sep 2020 11:10:51 +1000https://danmackinlay.name/post/reading_group_2020_causal_dags.htmlMotivational examples Generally Machinery Structural Equation Models Directed Acyclic Graphs (DAGs) Causal interpretation do-calculus Case study: Causal GPs Recommended reading Quick intros Textbooks Questions References See also a previous version, and the notebook on causal inference this will hopefully inform one day.
We follow Pearl’s summary (Pearl 2009a), sections 1-3.
In particular, the paper surveys the development of mathematical tools for inferring (from a combination of data and assumptions) answers to three types of causal queries: (1) queries about the effects of potential interventions, (also called “causal effects” or “policy evaluation”) (2) queries about probabilities of counterfactuals, (including assessment of “regret,” “attribution” or “causes of effects”) and (3) queries about direct and indirect effects (also known as “mediation”)Causal Bayesian networks
https://danmackinlay.name/notebook/causal_bayesian_networks.html
Tue, 01 Sep 2020 08:29:10 +1000https://danmackinlay.name/notebook/causal_bayesian_networks.htmlReferences Some kind of alternative graphical formalism for causal independence graphs 🤷?
discrete probability trees, sometimes also called staged tree models. A probability tree is one of the simplest models for representing the causal generative process of a random experiment or stochastic process The semantics are self-explanatory: each node in the tree corresponds to a potential state of the process, and the arrows indicate both the probabilistic transitions and the causal dependencies between them.Modern conspiracy mania
https://danmackinlay.name/notebook/conspiracy_mania.html
Mon, 17 Aug 2020 16:54:28 +1000https://danmackinlay.name/notebook/conspiracy_mania.htmlThis one is very much of the moment, and is an interesting weaponized social media strategy. I should dump the many links here.
QAnon as a nexus, model and test case. Charlotte Alter/Kenosha, in TIME conspiracies in the 2020 election
Aragorn Eloff argues that there is an informational-class war. Hm. See a timeline of the conspiracy theorising of the alt-right.
It is hip at the moment to consider the QAnon conspiracy network in particular in the context of collaborative games.Dilemmas of collective action
https://danmackinlay.name/notebook/collective_action.html
Sat, 01 Aug 2020 08:09:41 +1000https://danmackinlay.name/notebook/collective_action.htmlRabbit and stag game Coordination on conflict Collective action dilemmas for elites Trust References The challenges of coordination in collective action.
Cooperation problems writ large. How do a million people work together?
See also democracy etc.
Rabbit and stag game A game theoretical model of coordination which throws some light on the problems of social coordination. This is pertinent in problems of, e.g. having democracy or of getting peer networks off Facebook.Recommended workplace habits
https://danmackinlay.name/notebook/recommended_workplace_habits.html
Sun, 19 Jul 2020 11:41:52 +1000https://danmackinlay.name/notebook/recommended_workplace_habits.html Half way between ideal teamwork and moral mazes is are tips that are attainable by workers of low rank within an organisation who would like to do well for themselves and/or the organisation.
Brag documents keep your skills apparent Telling your colleague’s managers they did great Difficulty Anchoring to make sure your work is valued appropriately (esp advocated for under-represented minorities.) Designing less toxic social media
https://danmackinlay.name/notebook/kinder_social_media.html
Sun, 12 Jul 2020 07:36:30 +1000https://danmackinlay.name/notebook/kinder_social_media.htmlWhat if we could do less harm minimisation of social network behaviour because they were less toxic, had less addictiveness and were less fruitful for weaponized corrosivity?
What might such networks look like? Since this is an evolutionary process, I suspect we need to consider an ongoing design process rather than a terminal state. But starting points.
Nick punt on de-escalating twitter.
I am less persuaded by debudbble the ‘thoughtful twitter debate’ app.(Approximate) matrix factorisation
https://danmackinlay.name/notebook/matrix_factorisation.html
Fri, 03 Jul 2020 19:51:38 +1000https://danmackinlay.name/notebook/matrix_factorisation.htmlWhy does it ever work Overviews Non-negative matrix factorisations As regression Sketching \([\mathcal{H}]\)-matrix methods Randomized methods Connections to kernel learning Implementations References Forget QR and LU decompositions, there are now so many ways of factorising matrices that there are not enough acronyms in the alphabet to hold them, especially if you suspect your matrix is sparse, or could be made sparse because of some underlying constraint, or probably could, if squinted at in the right fashion, be such as a graph transition matrix, or Laplacian, or noisy transform of some smooth object, or at least would be close to sparse if you chose the right metric, or…Teamwork
https://danmackinlay.name/notebook/teamwork.html
Thu, 02 Jul 2020 07:20:47 +1000https://danmackinlay.name/notebook/teamwork.htmlReferences Two-pizza rules, diversity for efficiency, communication, the human dimension of project management.
I would like more quantified and peer-reviewed content here, but I will take what I can get.
Coda Hale, Work is work:
Keep the work parallel, the groups small, and the resources local. When presented with a set of problems which grow superlinearly intractable as \(N\) increases, our best bet is to keep \(N\) small.Economics of insurgence
https://danmackinlay.name/notebook/economics_of_insurgence.html
Wed, 01 Jul 2020 07:30:05 +1000https://danmackinlay.name/notebook/economics_of_insurgence.htmlReferences The sophisticated business strategy of modern insurgency. MBAs in terror. Business analytics for utopia. Social media strategy for the apocalypse.
The women march to Versailles, 1789
I do not know much about this but a few links of interest are here.
Stand Alone Complex is a handy word in this domain.
A ‘Stand Alone Complex' can be compared to the copycat behavior that often occurs after incidents such as serial murders or terrorist attacks.Research discovery
https://danmackinlay.name/notebook/research_discovery.html
Fri, 26 Jun 2020 09:27:43 +1000https://danmackinlay.name/notebook/research_discovery.htmlReading groups and co-learning Paper analysis/annotation Finding copies Recommender systems for academics are hard and in particular, I suspect they are harder than normal because definitionally the content should be new and hard to relate to existing stuff. Indeed, finding connections is a publishable result in itself.
Complicated interaction with systems of peer review. Could a normal recommender system such as canopy be made to work for academics?Learning of manifolds
https://danmackinlay.name/notebook/learning_of_manifolds.html
Tue, 23 Jun 2020 09:34:49 +1000https://danmackinlay.name/notebook/learning_of_manifolds.htmlImplementations TTK scikit-learn tapkee References 🏗🏗🏗🏗🏗
I will restructure learning on manifolds and dimensionality reduction into a more useful distinction.
Berger, Daniels and Yu on manifolds in Genome search
As in — handling your high-dimensional, or graphical, data by trying to discover a low(er)-dimensional manifold that contains it. That is, inferring a hidden constraint that happens to have the form of a smooth surface of some low-ish dimension.Economics on networks
https://danmackinlay.name/notebook/network_economics.html
Tue, 16 Jun 2020 09:01:47 +1000https://danmackinlay.name/notebook/network_economics.htmlReferences Charles-Joseph Minard’s classic economic network visualisation
Placeholder; The models in economics where the interactions happen over graphs, usually denoting some kind of relationship, e.g. between consumers or businesses. I do not have any particular knowledge of this area except for some incidental study when I was planning on doing a PhD in the area a long time ago.
For estimation theory of economic networks I would look under inference on social graphs.Soft methodology of science
https://danmackinlay.name/notebook/soft_methodology_of_science.html
Thu, 11 Jun 2020 11:15:12 +1000https://danmackinlay.name/notebook/soft_methodology_of_science.htmlMiscellany of career tips How precious is my idea? Sidling up to the truth. Disruption by field outsiders References The course of science
In which I collect tips from esteemed and eminent minds about how to go about pro-actively discovering stuff. More meta-tips than detailed agendas of discovery.
Miscellany of career tips A current meta-question. One starting point: John Schulman’s Opinionated Guide to ML Research, which discusses stuff like this:Science, history and philosophy thereof
https://danmackinlay.name/notebook/hps.html
Fri, 05 Jun 2020 09:34:20 +1000https://danmackinlay.name/notebook/hps.htmlWhat is science? References I do not myself have a much to say about the philosophy of science as such. I read a lot of Lakatos that one time.
Mostly I am interested in a kind of qualitative mechanism design musing as it pertains to designing better peer-review.
What is science? Not really in that vein, check out amusing curmudgeon: DC Stove, Popper and after: Four modern irrationalists.Project management
https://danmackinlay.name/notebook/project_management.html
Fri, 05 Jun 2020 08:01:43 +1000https://danmackinlay.name/notebook/project_management.htmlEmpirically informed Operations research style Skunk works style Solo Overruns The simplest thing References Making projects happen via a combination of institutional affordances and teamwork. I am especially interested in innovative/research/invention which by design have a different structure than something more reasonably routine and repeatable, such as building a house.
Empirically informed This field is full of MBAs citing anecdata. How much empirical statistically meaningfully study is there of efficient organisations actually?Pluralistic ignorance
https://danmackinlay.name/notebook/pluralistic_ignorance.html
Wed, 03 Jun 2020 18:45:31 +1000https://danmackinlay.name/notebook/pluralistic_ignorance.htmlReferences Placeholder. In social norms what if the norm is not what we all personally believe we should do, but rather waht we all believe we all believe we should do but in fact few of us believe
To discover:
Is that keyword I have seen floating about, the “spiral of silence”, the same thing? Rhetorical uses of claims to pluralistic ignorance. Relationship to majority illusions.Inference on social graphs
https://danmackinlay.name/notebook/inference_on_social_graphs.html
Wed, 03 Jun 2020 18:20:13 +1000https://danmackinlay.name/notebook/inference_on_social_graphs.htmlConnection to statistical relational learning Majority Illusions and filter bubbles Confounding on graphs To file References Placeholder.
Fun keywords: Egocentric sampling, friendship paradox, majority illusion, and the analysis of projectivity. 🏗
Connection to statistical relational learning I cannot help but notice that the discussions of changing probabilistic domain, and unusual assumptions about exchangeability/projectivity are reminiscent of inference on social graphs. Connections?
Majority Illusions and filter bubbles In homophilic networks (0.Natural gradient descent
https://danmackinlay.name/notebook/natural_gradient.html
Tue, 26 May 2020 14:29:16 +1000https://danmackinlay.name/notebook/natural_gradient.htmlNatural Policy Gradient References A placeholder.
The misalignednment of the magnetic and geographic poles is not an ideal metaphor for the misalignment of natural and euclidean gradients (because magnetic field descent leads to the wrong optimum, not the wrong rate) but I can’t be searching online for illustrations all day there is work to do.
Gradient descent with the natural gradient, or a close approximation thereto.Incentive mechanism design
https://danmackinlay.name/notebook/mechanism_design.html
Sun, 24 May 2020 12:01:20 +1000https://danmackinlay.name/notebook/mechanism_design.htmlExamples Tutorials References Mecha design, courtesy blueprintbox.
“Reverse” game theory, bargaining, auctions, pie slicing, swarm sensing for autonomous agents. Reputation mechanisms. Voting systems are mechanisms, and the Arrow impossibility theoryem and Gibber-Satterthwaite theorem are foundational results in mechanism design. The classic, assassination markets, are no longer at the vanguard according to Brian Merchant, but prediction markets are a classic incentive mechanism for distributing forecasts. (Merchant 2020) Every blockchain-style cryptowhatsit is a mechanism design problem.But what can I do?
https://danmackinlay.name/notebook/but_what_can_i_do.html
Tue, 19 May 2020 20:40:44 +1000https://danmackinlay.name/notebook/but_what_can_i_do.htmlBuild networks Contribute to organisations who make it better Fix the media “Death equalises sceptres and hoes”
It has come to my attention that many people feel that civilisation as we know it is under existential threat and that they are powerless to prevent it. I do not have much time to write this out at length at the moment, but I suggest humbly that powerlessness is itself a problem.Directed graphical models
https://danmackinlay.name/notebook/graphical_models_directed.html
Wed, 13 May 2020 13:40:29 +1000https://danmackinlay.name/notebook/graphical_models_directed.htmlReferences I found James F Fixx’s puzzle book on the shelf when writing this post
Graphs of conditional, directed independence are a convenient formalism for many models. These are also called Bayes nets (not to be confused with Bayesian inference.)
Once you have the graph, you can infer more detailed relations than mere conditional dependence or otherwise; this is precisely that hierarchical models emphasise.Contact tracing
https://danmackinlay.name/notebook/contact_tracing.html
Sun, 10 May 2020 07:21:14 +1000https://danmackinlay.name/notebook/contact_tracing.htmlDr. Evans, 1917, How to keep well
Privacy-respecting computing approaches are getting important in this time of epidemics.
A recent round up by Patrick Howell O'Neill, Tate Ryan-Mosley and Bobbie Johnson lists some of the apps in action.
John Langford:
For the following a key distinction to understand is between proximity and location approaches. In proximity approaches (such as DP3T, TCN, MIT PACT(*), Apple or one of the UW PACT(*) protocols which I am involved in) smartphones use Bluetooth low energy and possibly ultrasonics to discover other smartphones nearby.Learning with conservation laws, invariances and symmetries
https://danmackinlay.name/notebook/learning_with_conservation_laws.html
Fri, 01 May 2020 09:53:27 +1000https://danmackinlay.name/notebook/learning_with_conservation_laws.htmlReferences Failure of conservation of mass at system boundaries is a common problem in models with nonparametric likelihood
Learning in complicated systems where we know that there is a conservation law in effect. Or, more advanced, learning a conservation law that we did not know was in effect. As seen in especially ML for physics. This is not AFAIK a particular challenge in traditional parametric statistics where we can impose conservation laws on a problem through the likelihood, but nonparametrics models, or models with overparameterisation such as neural nets this can get fiddly.Statistical relational learning
https://danmackinlay.name/notebook/statistical_relational_learning.html
Mon, 27 Apr 2020 21:45:09 +1000https://danmackinlay.name/notebook/statistical_relational_learning.htmlReferences Placeholder.
I cannot help but notice that the discussions of changing probabilistic domain, and unusual assumptions about exchangability are reminiscent of inference on social graphs. Connections?
See the big book.
References Braz, Rodrigo de Salvo, Eyal Amir, and Dan Roth. 2008. “A Survey of First-Order Probabilistic Models.” In Innovations in Bayesian Networks, edited by Dawn E. Holmes and Lakhmi C. Jain, 156:289–317. Studies in Computational Intelligence.Statistical projectivity
https://danmackinlay.name/notebook/projectivity.html
Sun, 26 Apr 2020 19:23:30 +1000https://danmackinlay.name/notebook/projectivity.htmlReferences Placeholder.
Turns out that Cosma has been here very much first.
References Jaeger, Manfred, and Oliver Schulte. 2020. “A Complete Characterization of Projectivity for Statistical Relational Models.” April 23, 2020. http://arxiv.org/abs/2004.10984. Shalizi, Cosma Rohilla, and Alessandro Rinaldo. 2013. “Consistency Under Sampling of Exponential Random Graph Models.” Annals of Statistics 41 (2): 508–35. https://doi.org/10.1214/12-AOS1044. Snijders, Tom A. B. 2010. “Conditional Marginalization for Exponential Random Graph Models.Learning summary statistics
https://danmackinlay.name/notebook/learning_summary_statistics.html
Wed, 22 Apr 2020 14:40:07 +1000https://danmackinlay.name/notebook/learning_summary_statistics.htmlReferences A dimensionality reduction/feature engineering trick for likelihood-free inference methods such as indirect inference or approximate Bayes computation.
TBD. See de Castro and Dorigo (2019):
Simulator-based inference is currently at the core of many scientific fields, such as population genetics, epidemiology, and experimental particle physics. In many cases the implicit generative procedure defined in the simulation is stochastic and/or lacks a tractable probability density p(x|θ), where θ ∈ Θ is the vector of model parameters.Post stratification
https://danmackinlay.name/notebook/post_stratification.html
Wed, 22 Apr 2020 08:44:52 +1000https://danmackinlay.name/notebook/post_stratification.htmlReferences A trick for handling a non-random sampling problem particularly common in survey data.
MRP, a.k.a. Mister P, is one method for correcting for non-response bias and other suc bias sampling. See also RPP.
I have not used this tool practically and so am not at all qualified to comment.
What I can do is link to my reading list of examples and explainers:Survey modelling
https://danmackinlay.name/notebook/survey_modelling.html
Tue, 21 Apr 2020 19:26:06 +1000https://danmackinlay.name/notebook/survey_modelling.htmlSampling challenges Post stratification Cluster randomized trials Ordinal data Confounding and observational studies Graph sampling. Data sets References Tricks of particular use in modeling survey data. Hierarchical models to adjust for issues such as non random sampling and the varied great difficulties of eliciting human preferences by asking them. A grab bag of the weird data types, problems and sampling bias problems.
Sampling challenges What is that Lizardman constant?