428 Matching Annotations
  1. Last 7 days
    1. It also means that my annotations are in the paths of others and I need to consider that, forcing me to add context and consideration to my own notes.

      Epiphany! Social annotating while reading brings to READING a stance I try to have when WRITING: considering an audience. Do I read differently when my annotations mean my reading has an audience?

    1. reading fram

      The reading frame is the way of dividing each sequence of nucleotides in the DNA or RNA molecule into a set of consecutive, non-overlapping triplets.

  2. Oct 2020
    1. The idea of the hermeneutic circle is to envision a whole in terms how the parts interact with each other, and how they interact with the whole. That may sound a little bit out there, so let’s have a look at a concrete example.

      This is a general concept, the rest of the article extrapolates the idea to the act of reading. This may be a stretch, since it implies that whatever can be broken into parts will belong to the hermeneutic circle, while this only applies to interpreting (text)

    2. As objective you may try to be, interpreting a text doesn’t happen in a vacuum. The hermeneutic circle captures the complex interaction between an interpreter and a text.

      This is the only useful idea in the text. Whatever we read has the context in which it was written and the context in whcih it is being read. Is this a hermeneutic circle as described earlier? Don't think so.

    1. Quantum: Einstein, Bohr, and the Great Debate about the Nature of Reality Illustrated Edition by {"isAjaxComplete_B0032HW9M0":"0","isAjaxInProgress_B0032HW9M0":"0"} Manjit Kumar (Author) › Visit Amazon's Manjit Kumar Page Find all the books, read about the author, and more. See search results for this author Are you an author? Learn about Author Central Manjit Kumar (Author)
    1. Physics of the Impossible: A Scientific Exploration of the World of Phasers, Force Fields, Teleportation and Time Travel by {"isAjaxComplete_B000ARDFYQ":"0","isAjaxInProgress_B000ARDFYQ":"0"} Michio Kaku (Author) › Visit Amazon's Michio Kaku Page Find all the books, read about the author, and more. See search results for this author Are you an author? Learn about Author Central Michio Kaku (Author)
    1. How to Become a Straight-A Student: The Unconventional Strategies Real College Students Use to Score High While Studying Less Kindle Edition by {"isAjaxComplete_B001IGNR0U":"0","isAjaxInProgress_B001IGNR0U":"0"} Cal Newport (Author) › Visit Amazon's Cal Newport Page Find all the books, read about the author, and more. See search results for this author Are you an author? Learn about Author Central Cal Newport (Author)
    1. How to Be a High School Superstar: A Revolutionary Plan to Get into College by Standing Out (Without Burning Out) 1st Edition, Kindle Edition by {"isAjaxComplete_B001IGNR0U":"0","isAjaxInProgress_B001IGNR0U":"0"} Cal Newport (Author) › Visit Amazon's Cal Newport Page Find all the books, read about the author, and more. See search results for this author Are you an author? Learn about Author Central Cal Newport (Author)
    1. Rise of the Robots: Technology and the Threat of a Jobless Future Paperback – Illustrated, July 12, 2016 by {"isAjaxInProgress_B002RCTIHU":"0","isAjaxComplete_B002RCTIHU":"0"} Martin Ford (Author) › Visit Amazon's Martin Ford Page Find all the books, read about the author, and more. See search results for this author Are you an author? Learn about Author Central Martin Ford (Author)
    1. Soonish: Ten Emerging Technologies That'll Improve and/or Ruin Everything Hardcover – Illustrated, October 17, 2017 by {"isAjaxComplete_B06XHKDZVZ":"0","isAjaxInProgress_B06XHKDZVZ":"0"} Kelly Weinersmith (Author) › Visit Amazon's Kelly Weinersmith Page Find all the books, read about the author, and more. See search results for this author Are you an author? Learn about Author Central Kelly Weinersmith (Author), Zach Weinersmith
    1. The Future of Humanity: Our Destiny in the Universe Paperback – Illustrated, April 2, 2019 by {"isAjaxComplete_B000ARDFYQ":"0","isAjaxInProgress_B000ARDFYQ":"0"} Michio Kaku (Author) › Visit Amazon's Michio Kaku Page Find all the books, read about the author, and more. See search results for this author Are you an author? Learn about Author Central Michio Kaku (Author)
    1. Physics of the Future: How Science Will Shape Human Destiny and Our Daily Lives by the Year 2100 Paperback – Illustrated, February 21, 2012 by {"isAjaxComplete_B000ARDFYQ":"0","isAjaxInProgress_B000ARDFYQ":"0"} Michio Kaku (Author) › Visit Amazon's Michio Kaku Page Find all the books, read about the author, and more. See search results for this author Are you an author? Learn about Author Central Michio Kaku (Author)
    1. If you’d like to differentiate between the various functions a paragraph in a text can have, look out for signal words. For example, the following literal devices may indicate that the function is to build a mental model: schema, allegory, analogy, hypothesis, metaphor, representation, simile, theory. Put a corresponding “model” mark next to these.
    1. Luhmann didn’t only write a lot and developed the most complex of all theoretical bodies in the social sciences. He was known for his vast knowledge and deep thinking. He didn’t run to his Zettelkasten when you asked him something. This is because he practiced thinking through writing and processing in the context of the Zettelkasten.

      I read Zettelkasten (German for “slip box”, or “card index”) and immediately think commonplace book!

    2. The Barbell Method takes this into account by integrating your reading habit into your knowledge work with two steps: Read the book. Read swiftly but don’t skip any parts unless they make you vomit or put you to sleep. Mark all the passages that stand out and contain useful, interesting or inspiring information. Read the book a second time. But now you read the marked parts only. This time you make notes, connect them to past notes (Zettelkasten Method!) and think about what you’ve read. Make mindmaps, drawings, bullet points – everything that helps you to think more clearly.
    1. being able to follow links to “follow a conversation” that is threaded on Twitter.

      This is one of my favorite parts about my website and others supporting Webmention: the conversation is aggregated onto or more closely adjacent to the source. This helps prevent context collapse.

      Has anyone made a browser tool for encouraging lateral reading? I'd love a bookmarklet that I could click to provide some highly relevant lateral reading resources for any particular page I'm on.

    2. What’s lateral reading?

      lateral reading, noun

      When you open up the Hypothes.is sidebar on a website and read the highlights and annotations that provide additional context to the main context.

    1. It is worth asking why ebooks and e-readers like the Kindle treaded water after swimming a couple of laps. I’m not sure I can fully diagnose what happened (I would love to hear your thoughts), but I think there are many elements, all of which interact as part of the book production and consumption ecosystem.

      For me, and potentially for a majority of others, our memories have evolved to be highly location specific. It's far easier for me to remember what I've read when I read a physical book. I can often picture what I was reading at the top, middle, or bottom of the left or right page. This fact in addition to how far I am in the book gives me a better idea of where I am with respect to a text.

      These ideas are very subtle and so heavily ingrained in us that they're not very apparent to many, if at all.

      See also Knowledge and Power in Prehistoric Societies: Orality, Memory and the Transmission of Culture by Lynne Kelly (Cambridge University Press, 2015)

    1. why encourage posting before you’ve even read the thing? Because, at least my hope is, it’ll prevent posting a link from becoming an endorsement for the content at the other end of that link. There’s a natural tendency to curate what we associate with our online profiles and I think that’s, in large part, because we’ve spent a lot of time equating a user’s profile page with a user’s identity and, consequently, their beliefs. But I consume a wealth of content that I don’t necessarily agree with, and that helps to inform me, to shape my opinions, as much as the content that I agree with wholeheartedly.
    1. First of all, I wanted to learn more about how to inspire learners to read. And this means for me as an educator to create a technical and social environment that is welcoming and easy to participate in.
    1. be quick to start books, quicker to stop them, and read the best ones again right after you finish

      farnam street blog tips on reading

  3. Sep 2020
    1. There are four rules to Analytical Reading Classify the book according to kind and subject matter. State what the whole book is about with the utmost brevity. Enumerate its major parts in their order and relation, and outline these parts as you have outlined the whole. Define the problem or problems the author is trying to solve.

      the [[four rules to [[analytical reading]]]]

    2. At this point, you start to engage your mind and dig into the work required to understand what’s being said. I highly recommend you use marginalia to converse with the author.

      From the linked article

      Full ownership of a book only comes when you have made it a part of yourself, and the best way to make yourself a part of it— which comes to the same thing— is by writing in it.

      Really love that quote - the idea of [[marginalia]] is to write in the margins, take notes as you read - to ask questions and answer them, and context around the highlights.

      In turn, [[make the book your own]]

    3. Analytical reading is a thorough reading. If inspectional reading is the best you can do quickly, this is the best reading you can do given time.

      [[analytical reading]] is one of the [[four levels of reading]] - the goal is to get to know a particular text very well.

    4. Systematic skimming

      sub-type of [[inspectional reading]] - reading the table of contents, skimming pages, looking for the hooks - how does this relate to [[proximity principle]] and [[scanning patterns]] - but getting a sense enough to know "is this book worth adding to the collection" for deeper reading

    5. Inspectional Reading We’ve been taught that skimming and superficial reading are bad for understanding. That is not necessarily the case. Using these tools effectively can increase understanding. Inspectional reading allows us to look at the author’s blueprint and evaluate the merits of a deeper reading experience.

      I think this is where I've been able to use #ADHD to it's advantages at times is with [[inspectional reading]] - being able to skim a large amount of content and get a sense of what I want to dig into later on or not.

    6. Getting the Questions Clear — Rather than focus on the problems the author is trying to solve, you need to focus on the questions that you want answered.

      Knowing what you are looking for is important, usually when I dig into reading something - I do have questions in mind.

      I've also been littering my notes recently with #question tags - how can I make better use of these?

      I think this can help avoid the 'reading just for the sake of keeping busy' - if I know what I am looking for, but also at the same time - know if a book is worth reading, or one I should quit reading.

  4. Aug 2020
  5. Jul 2020
    1. It is important when reading online not tp pn;y read sources that conform to your point of view.

    1. Online Reading Comprehension

      Sharing what they've found with one another Students getting really excited when they've found something, want to show teacher Once students get one part, challenge them to find something new Evaluate the information; what features make it good? What is definition of best? Synthesize what you are finding; bring in all the information from different sources Multimodal ways of reading

    2. Online Reading Comprehension

      Put focus on learning coming from partners

    1. At least five processing practices occur during online research and comprehension, each requiring additional new skills and strategies when they take place online:

      5 Practices are:

      1. Identify important questions
      2. Locating information
      3. Evaluating information critically
      4. Synthesizing information
      5. Reading and writing to communicate
  6. Jun 2020
    1. But tagging, alone, is still not good enough. Even our many tags become useless if/when their meaning changes (in our minds) by the time we go retrieve the data they point to. This could be years after we tagged something. Somehow, whether manually or automatically, we need agents and tools to help us keep our tags updated and relevant.

      search engines usually can surface that faster (less cognitive load than recalling what and where you store something) than you retrieve it in your second brain (abundance info, do can always retrieve from external source in a JIT fashion)

    1. They face huge challenges,

      No shit. Just think of all the literacy modalities that someone younger has to deal with. Memes. Data visualizations. Tweets. Tiktok vids. Instagram stories. Digital annotation. Video annotation. YouTubes. Akkkk. Emojis. Gifs. Think of all the lateral communication modalities written, read, spoken that our students face daily. And the success with which most of them navigate these...well, it is extraordinary.

    1. what do we lose when we don’t read laterally, when we passively scroll information feeds and accept what seems to be true and dismiss what seems to be wrong.

      I read for me first. I read my way second. We all have reading blind spots. I think reading laterally reduces these blind spot, but it adds others. When I read people like Green it puts my back up and I say, "Who the fuck made you king of the lateral reading hill?" Another expert/sociopath rises up to keep me down. I teach to make people more powerful. Reading tools help, but the idiosyncratic ways we use those tools are grandly variably. I want it that way not one way.

  7. May 2020
    1. John Green in CrashCourse

      Since I only know John Green as a fiction writer, I'm going to link to a different lateral source named Mike Caulfield, a researcher and educator whose done some significant thinking and writing in the space: https://hapgood.us/?s=lateral+reading

    1. these words bring up all kinds of questions

      some thoughts when skimming through stream-of-consciousness journals like these

      if I want to absorb the information and "learn" faster, then reading faster or summarising the text is not the solution, because a text is already a compressed lossy encoded form of the initial thought. to decode it further and transfer it into my head would risk too much missing bits of information.

    1. Annotations—even inline marginalia which include your own writing—have very little informational value. They’re atomized; they don’t relate to each other; they don’t add up to anything; they’re ultra-compressed; they’re largely unedited. That’s fine: think of them as just a reminder. They say “hey, look at this passage,” with a few words of context to jog your memory about what the passage was about.Since you’re going to write lasting notes anyway, annotations need carry just enough information to recreate your mental context in that moment of reading. You wouldn’t want to rely on that long-term, since then you’d just have a huge pile of hooks you’d have to “follow” anytime you wanted to think about your experience with that book.

      Classical marginalia in books, according to Andy Matuschok, have little informational value. They are not interlinked, they're very compressed and usually unedited. But that's okay.

      Their purpose is to help you get back to the mental context you were in when you thought the passage was worth returning to.

  8. Apr 2020
    1. Regarding a Hunter S. Thompson book. The reviews on Amazon. One review said to read Fear and Loathing in Las Vegas first before this one inquiry mentions. So noted.

    1. 你說,這樣不是會忽略掉書中其他有價值的內容嗎?也許是,但是誰在乎?我是為了成長而學習,不是為了看完一本書而閱讀。在大學的時候,也沒有任何一堂課會上完了整本書,不是嗎?

      放棄完美主義

    1. 讀書e誌這個粉專的特色是,版主會分享許多的英文原文書,有許多是台灣根本還沒翻譯版的,這種新鮮感總是令我愛不釋手。書評的行文風格很親切易讀,篇幅不會太長,文章也同步會發表在 讀書e誌部落格 上。你可以選擇自己喜歡的平台來追蹤。

      不錯的資源

    1. There are good preprints and bad preprints, just like there are with journal articles. Overall, do not be afraid to be scooped or plagiarized! Preprints also actually protect against scooping [21,22]. Preprints establish the priority of discovery as a formally published item. Therefore, a preprint acts as proof of provenance for research ideas, data, code, models, and results—all outputs and discoveries.
      • Salah satu alasan untuk tidak mengunggah preprint adalah takut idenya dicuri,

      • Ini adalah faktor budaya yang lain. Ketakutan yang tidak beralasan. Justru dengan mengunggah preprint, peneliti dapat mengklaim ide lebih awal.

      • Preprint ada yang bagus dan ada yang buruk, peninjauan akan ada di tangan pembaca. Ini adalah hambatan budaya berikutnya, ketika mayoritas pembaca ingin melimpahkan tanggungjawab untuk memverifikasi, memeriksa, dan menjamin kualitas suatu makalah kepada para peninjau.

      • Pengalihan tanggungjawab ini sulit dilakukan ketika dokumen PR sendiri tertutup, dan tidak lepas dari bias.

      • Selain itu, dosen akan menyalahi prinsip yang disebarluaskan kepada para mahasiswa, untuk membaca secara kritis.

    1. Our phones and computers deliver unto each of us a personalized—or rather, algorithm-realized—distillation of headlines, anecdotes, jokes, and photographs. Even the ads we scroll past are not the same as our neighbor’s: a pair of boots has followed me from site to site for weeks. We call this endless, immaterial material a feed, though there’s little sustenance to be found.

      As reading materials—not just books, but newspapers, magazines, and ephemera—proliferated, more recent centuries saw the rise of reading “extensively”: we read these materials once, often quickly, and move on. Birkerts coins his own terms: the deep, devotional practice of “vertical” reading has been supplanted by “horizontal” reading, skimming along the surface. This shift has only accelerated dizzyingly in the time since Engelsing wrote in 1974, since Birkerts wrote in 1994, and since I wrote, yesterday, the paragraph above.

      Horizontal reading rules the day. What I do when I look at Twitter is less akin to reading a book than to the encounter I have with a recipe’s instructions or the fine print of a receipt: I’m taking in information, not enlightenment. It’s a way to pass the time, not to live in it. Reading—real reading, the kind Birkerts makes his impassioned case for—draws on our vertical sensibility, however latent, and “where it does not assume depth, it creates it.”

      We know perfectly well—we remember, even if dimly, the inward state that satisfies more than our itching, clicking fingers—and we know it isn’t here. Here, on the internet, is a nowhere space, a shallow time. It is a flat and impenetrable surface. But with a book, we dive in; we are sucked in; we are immersed, body and soul. “We hold in our hands a way to cut against the momentum of the times,” Birkerts assures. “We can resist the skimming tendency and delve; we can restore, if only for a time, the vanishing assumption of coherence. The beauty of the vertical engagement is that it does not have to argue for itself. It is self-contained, a fulfillment.”

    1. New media makes good literature (defined in relation to portrayal of character/human spirit) harder and as important as ever to find.

      Modern readers know more and more about psychology, and to them a writer’s explanations often seem unnecessary, false, or old-fashioned. They reads plenty, and no theme is shocking enough to surprise them. They get the facts from newspapers, magazines, radio, television, or movies. They’re connected to all the corners of the world — and nothing invented by the mind can compare with what takes place in reality. There’s still a chance that, in our day — or yours — humankind will reach the moon, or one of the planets. All the fantasies of so-called “science fiction” will pale in comparison with footage shot on the moon or on the other planets.

      The new era, in a sense, brought a radical change. Readers get endless information from the radio, from film, from the press, from television. They hear lectures on psychology and psychoanalysis. They watch travel shows, often enough travel themselves, and have knowledge of the world. If literary fiction and theater were to continue playing their old role, they would need an audience with a strong interest in human character and individuality, independent of all these other byproducts and external concerns. But the number of such connoisseurs is small. Real and pure connoisseurs of art are nearly as rare as real and pure artists.

      Precisely because people today are surrounded by a sea of information related to all kinds of fields, genuine modern artists have to deliver more and more artistic purity, more substance, a greater focus on the portrayal of character and individuality. But for this one has to have exceptional gifts. It is, simply put, harder than ever to be original and creative in new ways.

    1. Trying too hard to understand the minutiae initially can make it easy to miss a fairly apparently general theme. It’s productive to learn the parts that are accessible before putting in effort to fully understand everything. Don’t miss the forest for the trees.
    1. an evaluation of social reading platforms; an analysis of social reading applications;

      This book includes a few sections about defunct Readmill.

  9. Mar 2020
  10. Feb 2020
    1. The Four-Component Instructional Model (4C/ID) immediately grabbed my attention when looking at the various models we had available. The process involves (a) learning tasks, (b) part-task practice, (c) support information, and (d) procedural information. It seems simple enough but design guidelines laid out in the tables within Merriênboer (2019) provide very thorough suggestions and a constructivist approach. The end product for a curricula or course designed in this approach is one with fidelity to professional activity.

    2. Table 4:

      encourages metacognition and challenges conceptual and structural models.

    3. Table 3

      Encourages individualization, personalized learning tasks and self-directed learning. Shifts the responsibility over task selection from the teacher to the student (important!)

    4. Table 2:

      It's asking for thorough standards of acceptable performance to be set down, or made concrete by the instructor and ID, as a way to measure things effectively.

    5. Table 1: Design Principles for Learning Tasks

      one of the things I like about this model is that it is asking you from the get go to think of profession-related activities that can be designed or simulated for students to understand and relate the material.

    6. Instructional methods for part-task prac-tice aim at the strengthening of cognitive rules by extensive repetitive practice. Strengthening is a basic learning process that ultimately leads to fully automated cognitive schemas (Anderson, 1993

      Find the activities in a college class that require routines and rote learning, or cognitive schemas that require practice and "strengthen them"

    7. Procedural information (in Figure 1, the yellow beam with arrows pointing upwards to the learning tasks) helps students with performing the routine aspects of learning tasks, that is, aspects that are always per-formed in the same fashion.

      so break down a lecture theory session from a step by step session. Or a document that has theory information should have a clear divide when you get into the step by step process?

    8. sup-portive information in Figure 1 is not con-nected to individual learning tasks but to levels of complexity; it can be presented be-fore learners start to work on the learning tasks (under the motto ‘first the theory and only then start to practice’) and/or it can be consulted by learners who are already working on the learning tasks (under the motto ‘only consult the theory when needed’)

      The idea of providing this supplemental information as a way to check against their current cognitive models by referring to theory or let theory guide their attempt at a learning task. It doesn't dictate a rigid structure.

    9. This feedback stimulates learners to critically compare their own mental models and cognitive strategies with those of others, including experts, teachers and peer learners

      I like that this model includes the idea of feedback and how to trigger thinking and reflection on their own behavior, cognitive models, etc.

    Tags

    Annotators

    1. * Information-Processing Analysis : about the mental operations used by a person who has learned a complex skills

      this sounds a lot more involved unless you are working off a basic set of assumptions for mental operations and complex skills. Further understanding of psychological research and learning theories would be needed.

    2. Dick and Carey Model

      what the heck is this website, lmao.

    1. (a) learning tasks, (b) supportive information, (c) procedural information, and (d) part-task practice

      4 components

      Learning Tasks:

      • aim at integration of (non-recurrent and recurrent) skills.
      • provide authentic, whole-task experiences based on real-life tasks
      • Are organized in simple-to-complex task classes and have diminishing support in each task class (scaffolding).
      • Show high variability of practice.
    2. Learning tasks provide the backbone of the educational program; they provide learning from varied experiences and explicitly aim at the transfer of learning.

      the backbone of this model

    3. four-component instructional design (4C/ID)

      4C/ID tag might be one of the models I want to look into as I might use it for the Art Modules Class

    1. Feedback is a constant loop and not something that should come right at the end of the e-learning course or training module. So, it is important for courses to have feedback inserted at the right places during instruction.

      feedback is a constant loop

    2. Feedback needs to be specific in nature, as well as confirmatory and corrective. This way the learners know what they did right and wrong.

      how to give feedback

    3. Giving learners an indication of the desired outcomes help them calibrate their approach appropriately. This can be done via examples, case studies, and modelling various learning strategies like concept mapping, visualizing, role playing.

      learner guidance

    4. because it establishes some important relationships and concepts

      another link to article. this dude needs to elaborate more even if it makes the article longer, wtf!

    5. Michael Hanley, discusses this very subject in his article

      link to article

    6. Dale H. Schunk, in his book Learning Theories

      another book rec

    7. Successive Approximation Model (SAM)

      first time i've heard of this one

    8. Understanding Instructional Design

      reading check 3 material

  11. Jan 2020
    1. Suppose the algorithm chooses a tree that splits on education but not on age. Conditional on this tree, the estimated coefficients are consistent. But that does not imply that treatment effects do not also vary by age, as education may well covary with age; on other draws of the data, in fact, the same procedure could have chosen a tree that split on age instead

      a caveat

    2. hese heterogenous treatment effects can be used to assign treatments; Misra and Dubé (2016) illustrate this on the problem of price targeting, applying Bayesian regularized methods to a large-scale experiment where prices were randomly assigned

      todo -- look into the implication for treatment assignment with heterogeneity

    3. Chernozhukov, Chetverikov, Demirer, Duflo, Hansen, and Newey (2016) take care of high-dimensional controls in treatment effect estimation by solving two simultaneous prediction problems, one in the outcome and one in the treatment equation.

      this seems similar to my idea of regularizing on only a subset of the variables

    4. These same techniques applied here result in split-sample instrumental variables (Angrist and Krueger 1995) and “jackknife” instrumental variables

      some classical solutions to IV bias are akin to ML solutions

    5. Understood this way, the finite-sample biases in instrumental variables are a consequence of overfitting.

      traditional 'finite sample bias of IV' is really overfitting

    6. Even when we are interested in a parameter β ˆ, the tool we use to recover that parameter may contain (often implicitly) a prediction component. Take the case of linear instrumental variables understood as a two-stage procedure: first regress x = γ′z + δ on the instrument z, then regress y = β′x + ε on the fitted values x ˆ. The first stage is typically handled as an estimation step. But this is effectively a prediction task: only the predictions x ˆ enter the second stage; the coefficients in the first stage are merely a means to these fitted values.

      first stage of IV -- handled as an estimation problem, but really it's a prediction problem!

    7. Prediction in the Service of Estimation

      This is especially relevant to economists across the board, even the ML skeptics

    8. New Data

      The first application: constructing variables and meaning from high-dimensional data, especially outcome variables

      • satellite images (of energy use, lights etc) --> economic activity
      • cell phone data, Google street view to measure wealth
      • extract similarity of firms from 10k reports
      • even traditional data .. matching individuals in historical censuses
    9. Zhao and Yu (2006) who establish asymptotic model-selection consistency for the LASSO. Besides assuming that the true model is “sparse”—only a few variables are relevant—they also require the “irrepresentable condition” between observables: loosely put, none of the irrelevant covariates can be even moderately related to the set of relevant ones.

      Basically unrealistic for microeconomic applications imho

    10. First, it encourages the choice of less complex, but wrong models. Even if the best model uses interactions of number of bathrooms with number of rooms, regularization may lead to a choice of a simpler (but worse) model that uses only number of fireplaces. Second, it can bring with it a cousin of omitted variable bias, where we are typically concerned with correlations between observed variables and unobserved ones. Here, when regular-ization excludes some variables, even a correlation between observed variables and other observed (but excluded) ones can create bias in the estimated coefficients.

      Is this equally a problem for procedures that do not assum sparsity, such as the Ridge model?

    11. 97the variables are correlated with each other (say the number of rooms of a house and its square-footage), then such variables are substitutes in predicting house prices. Similar predictions can be produced using very different variables. Which variables are actually chosen depends on the specific finite sample.

      Lasso-chosen variables are unstable because of what we usually call 'multicollinearity.'<br> This presents a problem for making inferences from estimated coefficients.

    12. Through its regularizer, LASSO produces a sparse prediction function, so that many coefficients are zero and are “not used”—in this example, we find that more than half the variables are unused in each run

      This is true but they fail to mention that LASSO also shrinks the coefficients on variables that it keeps towards zero (relative to OLS). I think this is commonly misunderstood (from people I've spoken with).

    13. One obvious problem that arises in making such inferences is the lack of stan-dard errors on the coefficients. Even when machine-learning predictors produce familiar output like linear functions, forming these standard errors can be more complicated than seems at first glance as they would have to account for the model selection itself. In fact, Leeb and Pötscher (2006, 2008) develop conditions under which it is impossible to obtain (uniformly) consistent estimates of the distribution of model parameters after data-driven selection.

      This is a very serious limitation for Economics academic work.

    14. First, econometrics can guide design choices, such as the number of folds or the function class.

      How would Econometrics guide us in this?

    15. These choices about how to represent the features will interact with the regularizer and function class: A linear model can reproduce the log base area per room from log base area and log room number easily, while a regression tree would require many splits to do so.

      The choice of 'how to represent the features' is consequential ... it's not just 'throw it all in' (kitchen sink approach)

    16. Ta b l e 2Some Machine Learning Algorithms

      This is a very helpful table!

    17. Picking the prediction func-tion then involves two steps: The first step is, conditional on a level of complexity, to pick the best in-sample loss-minimizing function.8 The second step is to estimate the optimal level of complexity using empirical tuning (as we saw in cross-validating the depth of the tree).

      ML explained while standing on one leg.

    18. egularization combines with the observability of predic-tion quality to allow us to fit flexible functional forms and still find generalizable structure.

      But we can't really make statistical inferences about the structure, can we?

    19. This procedure works because prediction quality is observable: both predic-tions y ˆ and outcomes y are observed. Contrast this with parameter estimation, where typically we must rely on assumptions about the data-generating process to ensure consistency.

      I'm not clear what the implication they are making here is. Does it in some sense 'not work' with respect to parameter estimation?

    20. In empirical tuning, we create an out-of-sample experiment inside the original sample.

      remember that tuning is done within the training sample

    21. Performance of Different Algorithms in Predicting House Values

      Any reason they didn't try a Ridge or an Elastic net model here? My instinct is that these will beat LASSO for most Economic applications.

    22. We consider 10,000 randomly selected owner-occupied units from the 2011 metropolitan sample of the American Housing Survey. In addition to the values of each unit, we also include 150 variables that contain information about the unit and its location, such as the number of rooms, the base area, and the census region within the United States. To compare different prediction tech-niques, we evaluate how well each approach predicts (log) unit value on a separate hold-out set of 41,808 units from the same sample. All details on the sample and our empirical exercise can be found in an online appendix available with this paper athttp://e-jep.org

      Seems a useful example for trying/testing/benchmarking. But the link didn't work for me. Can anyone find it? Is it interactive? (This is why I think papers should be html and not pdfs...)

    23. Making sense of complex data such as images and text often involves a prediction pre-processing step.

      In using 'new kinds of data' in Economics we often need to do a 'classification step' first

    24. The fundamental insight behind these breakthroughs is as much statis-tical as computational. Machine intelligence became possible once researchers stopped approaching intelligence tasks procedurally and began tackling them empirically.

      I hadn't thought about how this unites the 'statistics to learn stuff' part of ML and the 'build a tool to do a task' part. Well-phrased.

    25. In another category of applications, the key object of interest is actually a parameter β, but the inference procedures (often implicitly) contain a prediction task. For example, the first stage of a linear instrumental variables regres-sion is effectively prediction. The same is true when estimating heterogeneous treatment effects, testing for effects on multiple outcomes in experiments, and flexibly controlling for observed confounders.

      This is most relevant tool for me. Before I learned about ML I often thought about using 'stepwise selection' for such tasks... to find the best set of 'control variables' etc. But without regularisation this seemed problematic.

    26. Machine Learning: An Applied Econometric Approach

      Shall we use Hypothesis to have a discussion ?

  12. Dec 2019
  13. Nov 2019
    1. Cabinet: one author or several; posts curated into particular collections or series’, often with thematic groupings, perhaps a “start here” page for new readers, or other pointers to specific reading sequences

      Colin Walker has suggested something like this in the past and implemented a "required reading" page on his website.

    1. For those not familiar with GPT-2, it is, according to its creators OpenAI (a socially conscious artificial intelligence lab overseen by a nonprofit entity), “a large-scale unsupervised language model which generates coherent paragraphs of text.” Think of it as a computer that has consumed so much text that it’s very good at figuring out which words are likely to follow other words, and when strung together, these words create fairly coherent sentences and paragraphs that are plausible continuations of any initial (or “seed”) text.

      This isn't a very difficult problem and the underpinnings of it are well laid out by John R. Pierce in An Introduction to Information Theory: Symbols, Signals and Noise. In it he has a lot of interesting tidbits about language and structure from an engineering perspective including the reason why crossword puzzles work.

      close reading, distant reading, corpus linguistics

    1. From this page:

      AUPresses thinks more readers should be aware of the work they’re doing. That’s why during the organization’s annual University Press Week, it launched a reading list it’s calling READ. THINK. ACT., a list of 75 peer-reviewed books designed to help non-academic readers understand the world and work to make it a better place.

    1. Sustained reading of long form texts and books is perhaps the most “active” of all basic media consumption. Philip Roth, in 2009, prognosticating the death of the novel, smartly points out: To read a novel requires a certain amount of concentration, focus, devotion to the reading. If you read a novel in more than two weeks you don’t read the novel really. So I think that kind of concentration and focus and attentiveness is hard to come by – it’s hard to find huge numbers of people, large numbers of people, significant numbers of people, who have those qualities. I love this: “If you read a novel in more than two weeks you don’t read the novel really.” Meaning: To truly read (and, I might add, write) is to commit and maintain focus long enough to live fully within the world of the book (as opposed to ten second dips in and out, as we mostly do with much online media).

      Craig Mod on really reading. This has me thinking about Philip Glass' discussion at the end of Words with Music where he talks about different worlds.

  14. Oct 2019
    1. when reading we can double back and re-read a word several times until it sticks (which according to some arguments may actually impair comprehension rather than increase it),
  15. Sep 2019
    1. Goodreads is nearly useless for finding recommendations

      I believe that the point of Goodreads—since Amazon bought the site—is lost here.

      The point of Goodreads is to make people buy books from Amazon. They're capitalists. They don't care about the common good, or about making people find books that they can truly benefit from.

  16. Aug 2019
    1. Both artists, through annotation, have produced new forms of public dialogue in response to other people (like Harvey Weinstein), texts (The New York Times), and ideas (sexual assault and racial bias) that are of broad social and political consequence.

      What about examples of future sorts of annotations/redactions like these with emerging technologies? Stories about deepfakes (like Obama calling Trump a "dipshit" or the Youtube Channel Bad Lip Reading redubbing the words of Senator Ted Cruz) are becoming more prevalent and these are versions of this sort of redaction taken to greater lengths. At present, these examples are obviously fake and facetious, but in short order they will be indistinguishable and more commonplace.

  17. Jul 2019
    1. In Reader, Come Home: The Reading Brain in a Digital World, Maryanne Wolf talks about how technology has led to more skimming rather than reading slowly and carefully. She talks about the benefits of “cognitive patience.” And she reminds us that reading quickly isn’t what makes someone a good reader.
  18. Jun 2019
    1. tl;dr: understanding reading styles for better reading

      Approach: An outline, high-level overview approach is taken since the article is so clearly structured. Specific ideas are highlighted and annotated.

      Outline

      Reading optimally means applying the suitable reading method for the suitable reading material you've selected, to fit the object you're trying to accomplish.

      Objectives:

      • Information vs Understanding
      • All the true value & growth is in understanding
      • Heuristic for information: anything easily digested is information.

      Reading Styles:

      1. Inspectional Reading: systemic skimming and superficial reading
      2. Analytical Reading:

      Rules:

      • Classify the book according to kind and subject matter
      • tl;dr
      • outline
      • defining problems/issues
      1. Syntopical Reading:

      Steps:

      • find relevant passages
      • understand in your own context
      • getting the questions clear
      • defining the issue
      • analyzing the issue
    1. Why I'm reading this: because I'm going to Refactor Camp & I figured I should be up on Ribbonfarm stuff.

  19. May 2019
  20. Apr 2019
    1. vocabulary notes

      interpreting plot

      • title
      • subtitle
      • climax
      • denouement
      • exposition
      • frame narrative
      • in media res

      summary

      • introduction
      • main body
      • denouement
      • conclusion

      literaty terms

      • first person narrative
      • second person narrative
      • third person narrative
      • irony
      • satire
      • epithet
      • personification
  21. Mar 2019
    1. Two-fold arguments are also put forward concerning the just and the unjust

      Examples here are more oriented toward personal freedom / libertarian values; points out that the same action applied to the same person, can be either potentially just or unjust depending on the circumstances, and also that there will be local conflicts and differences in opinion (e.g. in preventing a friend from committing suicide, they may be angry and disagree that your actions are just, though others may support your decision and think it is indeed just).

      There is no universal sanctity of property rights or freedom from bodily restraint by others; violence is in some cases justified, and in others not.

    2. Since if anyone should ask those who say that the same thing is both disgraceful and seemly whether they have ever done anything seemly, they would admit that they have also done something disgraceful, if disgraceful and seemly are really the same thing.

      In removing the context, the actions become effectively neutral, in that they are simultaneously good and bad; the actions themselves exist without judgment, and the judgment is only the product of the culture in which they take place (or are regarded)

    3. but the right occasion

      The right occasion = the opportune moment; context-specific.

    4. against the law

      As examples rise in degree of 'extremity', they brush up against norms and laws, an act that may be celebrated in one culture is punishable by death or ostracism in another, simply based on the geography or time in which it takes place. By keeping location and temporality intact, the author is able to refrain from making absolute claims about any of the actual behaviors and just cite them as things that are, irrespective of judgment. The degree to which the reader judges them may be dependent upon the reader's interpreting the behaviors not as context-specific acts, but as archetypes(?) or fixed representations of those acts which stand outside of time and place.

    5. the most beautiful grave imaginable

      Shifts to more 'extreme' examples, but points out that this perverse (to the greeks) act is BEAUTIFUL and an act of love to others; perhaps their reverence is inversely proportional to the Greeks' horror.<br> Also, in using such 'extreme' examples, the author shows that in fact nothing is truly extreme, because it is all a matter of context, and concepts of extremity introduce limits or constrain these things to a spectrum which is not necessarily accurate; it all depends on the context, and something cannot "depend" strongly or weakly based on the actual act, but only on the context.

    6. I go on to the things which cities and peoples regard as disgraceful

      Switches to point out arbitrary differences in culture; be born in one area and you believe x, be born in another and you believe y, but largely it is a matter of the random happenstance of one's birth. These beliefs are human creations, and vary depending on where the humans live.

    7. (although for men to do so in the palaistra aid gymnasium is seemly.)

      The "good" and the "bad" can be seemingly arbitrarily different between identity groups. Why is it seemly for person of type x and unseemly for person of type y?

    8. And I am not saying what the good is, but I am trying to explain that the bad and the good are not the same but that each is distinct from the other

      Not trying to identify a moral absolute, just point out that it is relative and therefore that there is no absolute.

    9. I think it would not be clear what was good and what was bad if they were just the same and one did not differ from the other; in fact such a situation would be extraordinary

      Is this sarcasm? Is he saying that such a duality would be extraordinary in that it would violate the philosophers' attempts to categorize and assign general rules? Not sure...

    10. But there is another argument which says that the good is one thing and the bad another, and that as the name differs, so does the thing named.

      This serves as a sort of refrain in the verse/chorus structure of the text. It is constructed like a song in some respects.

    11. And death is bad for those who die but good for the undertakers and gravediggers.

      Use of the progressive method of providing examples; in this case, linear from health to death. Further on in the text from mundane to extreme.<br> Examples here shift from the personal (the sick individual) to a class (professions); there is a hierarchy and blending here of sorts in that any member of any of the professions listed could find themselves as the individual afflicted by the example condition/problem, and as such we find that the same person could potentially hold these conflicting opinions at different stages in their life, and that neither is necessarily wrong nor contradictory.

    12. or at one time good and at another time bad for the same person

      Not fixed

    13. Suppose someone should question the man who says this as follows: Why don't you assign your household slaves their tasks by lot, so that if the teamster drew the office of cook, he would do the cooking and the cook would drive the team, and so with the rest ?

      How do these "fish out of water" statements compare back to previous examples? Seems to imply that, if you took an Athenian and placed them in Sparta, that they would consider the Spartan culture still foreign and would be at a disadvantage trying to operate within the context (which is likely)? They would see things through the lens of an Athenian, which, on the other hand, may provide certain perspective that the Spartans take for granted. Perhaps it is a reminder that opportunity is not democratically distributed, and that the moments and circumstances conducive to certain results cannot be manufactured by moving the pieces around, because they depend so much not only on the context in which they happen, but the experience and history of those who find themselves within the situation?