77 Matching Annotations
  1. Aug 2019
    1. Em 2015, o serviço de streaming de música Spotify criou a playlist chamada Descobertas da Semana, que funciona como uma curadoria digital. O algoritmo responsável por esta playlist utiliza técnicas de Filtragem Colaborativa, Processamento de Linguagem Natural e Processamento de Sinais de Áudio através de Redes Neurais Convolucionais para compor a playlist semanalmente.[33]
  2. Mar 2019
  3. Sep 2017
    1. Theorem 1.1.LetMn1andMn2be hypersurfaces ofNn+1that are tan-gent atpand let0be a unitary vector that is normal toMn1atp. SupposethatMn1remains aboveMn2in a neighborhood ofpwith respect to0. De-note byH1r(x)andH2r(x)ther-mean curvature atx2WofMn1andMn2,respectively. Assume that, for somer,1rn, we haveH2r(x)H1r(x)in a neighborhood of zero; ifr2, assume also that2(0), the principal cur-vature vector ofM2at zero, belongs tor. ThenMn1andMn2coincide in aneighborhood ofp

      Princípio da tangência no interior, para as curvaturas médias de ordem superior.

    2. A TANGENCY PRINCIPLE AND APPLICATIONS 215Suppose thatMn1remains aboveMn2in a neighborhood ofpwith respect to0.Denote byH1r(x)andH2r(x)ther-mean curvatures atx2WofMn1andMn2,respectively. Assume that, for somer,1rn, we haveH2r(x)H1r(x)in a neighborhood of zero. Ifr2, assume also that2(0), the principalcurvature vector ofM2at zero, belongs tor. ThenMn1andMn2coincide ina neighborhood ofp.

      Princípio da tangência no bordo, para as curvaturas médias de ordem superior.

  4. Jul 2017
    1. ξi(t) =ξip+t Xi−t22ΓijkXjXk+O(‖tX‖3)

      Usa-se o fato de que geodésicas são soluções do problema de valor inical:

      $$ \begin{aligned} \ddot{\gamma}^i_{p,q}(t) & = - \Gamma^i_{jk}(t) \dot{\gamma}^j_{p,q}(t) \dot{\gamma}^k_{p,q}(t) \\ \dot{\gamma}_{p,q}(0) & = X(p,q) \end{aligned} $$

    2. he canonical divergence D induces the metric g and the connections∇and∇∗. The same holds for the mean canonical divergence D∇mcd
    3. if∇is integrable, then it is notgenerally true that X(q,p) =−gradqD∇mcd(p‖·)
    4. mean canonical divergenceD∇mcd(p‖q):=12(D(p‖q) +D∗(q‖p))(64)which obviously satisfiesD(∇∗)mcd(p‖q) =D∇mcd(q‖p)
    5. he energy of the geodesicγp,qas the symmetrized version of the canonical divergence:12(D(p‖q) +D(q‖p))=12∫10∥∥ ̇γp,q(t)∥∥2dt

      Fazendo a mudança de variável $$ s \mapsto t(s) = 1- s $$ e usando o fato de que \( \gamma_{q,p}(s) = \gamma_{p,q}(1 - s) \), temos: $$ \begin{aligned} D(q||p) & := \int0^1 s ||\dot{\gamma}{q,p}(s)||^2 ds \ & = - \int1^0 (1 - t) ||\dot{\gamma}{p,q}(t)||^2 dt \ & = \int0^1 ||\dot{\gamma}{q,p}(t)||^2 dt - D(p||q) \end{aligned} $$

    6. D(p‖q) =∫10t∥∥ ̇γp,q(t)∥∥2dt(61)whereγp,qdenotes the geodesic from p to q.

      Até o momento, a conexão dual parece não desempenhar nenhum papel.

    7. D(p‖q) =∫10(1−t)∥∥ ̇γq,p(t)∥∥2dt
    8. inverse exponential map atγq,p(t)satisfiesXt(q,p)= (1−t) ̇γq,p(t)

      $$ \tilde{\gamma}_{\gamma_{q,p}(t),p}(s) = \gamma_{q,p}(t + s(1-t)), s \in [0,1] $$

      $$ \Longrightarrow X_t(q,p) := \dot{\tilde{\gamma}}_{\gamma_{q,p}(t),p}(s)\vert_{s=0} = (1 - t) \dot{\gamma}_{q,p}(t) $$

    9. n-dimensional dual manifold(M,g,∇,∇∗). Consider a∇-geodesicγq,p:[0, 1]→Mconnectingqandp. We define a tangent vector fieldXt(p,q)along this geodesic:Xt(q,p):=X(γq,p(t),p)(52)Obviously,X0=X(q,p)(53)X1(q,p) =0(54)Definition 3.A canonical divergence from p to q is defined by the path integralD(p‖q) =∫10〈Xt(q,p), ̇γq,p(t)〉dt

      Qual o papel da conexão dual?

    10. ∫10〈X(γ(t),p), ̇γ(t)〉dt=−∫10〈gradγ(t)Dp, ̇γ(t)〉dt=−∫10(dγ(t)Dp)( ̇γ(t))dt=−∫10d Dp◦γd t(t)dt=Dp(γ(0))−Dp(γ(1))=Dp(q)−Dp(p) =Dp(q) =D(p‖q)(13)In particular, we can apply this derivation to the geodesic connectingqandpeven when theintegrability ofXis not guaranteed and obtain the definition of a general canonical divergence
    11. functionsDpsatisfying the condition of Equation (12) then they are uniqueup to a constant that can vary withp, and we can therefore assumeDp(p) =0
    12. aRiemannian metricgonM. Given such a metric, we assumeintegrabilityofXand∇, respectively,in the sense that for allpthere exists a functionDpsatisfyingX(q,p) =−gradqDp
    13. although being quite restrictive in general, thisproperty will be satisfied in our information-geometric context, wheregis given by the Fisher metricand∇is given by them- ande-connections and their convex combinations, theα-connections
    14. pointspandq, one can interpret anyXwith expq(X) =pas a difference vectorXthattranslatesqtop
    15. p−q=−gradqDp(9)Here, the gradient gradqis taken with respect to the canonical inner product onRn

      De outra forma, podemos postular que a divergência canônica é a solução da edp: $$ D_p(q) = {1 \over 2 } |grad_q D_p|^2 $$

    16. fixed pointp∈M, we want to define a vector fieldq7→X(q,p), at least in a neighbourhood ofp, thatcorresponds to the difference vector field
    17. manifold is dually flat, a canonical divergence was introduced by Amari and Nagaoka [2], which isa Bregman divergence
    18. a divergence exists for any such manifold. However, it isnot unique and there are infinitely many divergences that give the same geometrical structure
    19. find a divergenceDwhich generates a given geometrical structure(M,g,∇,∇∗)
    20. the coefficientsDΓijk(p) =−∂i∂j∂′kD(ξp‖ξq)∣∣q=p(5)DΓ∗ijk(p) =−∂′i∂′j∂kD(ξp‖ξq)∣∣∣q=p(6)define a pair of dual affine connectionsD∇andD∇∗[1]. The duality of the connections holds withrespect to the Riemannian metricDgin terms of the following condition:X〈Y,Z〉=〈D∇XY,Z〉+〈Y,D∇∗XZ〉(7)for all vector fieldsX,YandZ, where the brackets〈·,·〉denote the inner product with respect toDg
    21. he coefficients of the Riemannian metric can be written asDgij(p) =−∂i∂′jD(ξp‖ξq)∣∣∣q=p=∂′i∂′jD(ξp‖ξq)∣∣∣q=p
    22. When a coordinate systemξ:p7→ξp= (ξ1p, . . . ,ξnp)∈Rnis given inM, we pose one condition that, for two nearby pointsξpandξq=ξp+∆ξ,Dis expanded asD(p‖q) =12Dgij(p)∆ξi∆ξj+O(‖∆ξ‖3)(2)and(Dgij(p))ijis a positive definite matrix.
    23. A divergence functionD(p‖q)is a differentiable real-valued function of two pointspandqin amanifoldM. It satisfies the non-negativity conditionD(p‖q)≥0(1)with equality if and only ifp=q.

      A saturação (rigidez) da desigualdade é uma espécie de não-degenerescência da divergência.

  5. Apr 2017
  6. Mar 2017
  7. Feb 2017
  8. Oct 2015
    1. being able to utilize both of them

      That's right -- they should be complement if at all possible.

    2. IF

      tell the reader what this means

    3. Social Science Journals

      Do you mean the Social Sciences Databases?

    1. The databases included in EBSCOhost are well-respected and reputable. A vast majority of articles in EBSCOhost are peer-reviewed scholarly journals. ProQuest is considered a world leader in serving the information needs of researchers of all ages and abilities in libraries and institutions across the globe.

      Here you can speak to how long they've been in business, how many libraries use them, as well as how well they represent the works they index.

    2. are part of the Kentucky Virtual Library database collection

      You're not the first person to bring this up --- I know about KVL, but where do you see this in our version of EBSCO?

    1. both the EBSCO and ProQuest databases

      EBSCO needs to provide a better interface for this, though.

    1. The ebrary e-books database through ProQuest covers a large range of subject areas,

      There's no browsing option, though, and subject headings are missing. These ebooks need to be available through UK's OPAC.

    2. About a sixth of the databases available in EBSCOhost are H.W. Wilson

      This merger isn't that old. I remember the old HW Wilson databases and interface.

    1. ProQuest also offers a basic search box at the top of the page, with options to select “Full text” and “Peer reviewed” sources, which is handy. However, using ProQuest’s databases for information seeking is probably best for browsing and less for known item searching. There is no general category for business or related topics, and the best approximation for searching for an example query like “statistics” is the “Social Sciences” category in the synopsis of the “PAIS International” database.  In doing a keyword search for “statistics” on the homepage, many of the results were from the “Health & Life Sciences” category, but no connection on the general topic. The equivalent starting place for “statistics” on EBSCO would be the “Business Source Complete”, showing that EBSCO has more favorable subject specificity for known item searching.

      See comment in Canvas about this.

    2. EBSCO seems to cater more towards advanced research in characteristics that will be explored in this paper

      bold claim!

    1. the same seems to be true for EBSCOhost

      there's documentation on this

    1. A benefit of Social Services Abstracts is that it also lets you see documents with shared references

      this is important

    2. Example of a command line search for these databases: SU(abuse) AND SU(art therapy)  AND TI(child)

      How does this query turn out in either database?

    3. I was able to locate the article in physical format to confirm EBSCOhost’s details and was able to assess that the particular entry I was looking for contained accurate information.

      Nice test case.

    1. being directed to more articles by the author

      Do you think it would be able to capture variations on names? E.g.:

      • Last, First
      • Last, First M
      • Last, F. M.
    1. which may mean there needs to be a broader search conducted

      Yes, searches are not immediately transferable, and need to be translated.

    2. Both EBSCOhost and ProQuest offer a variety of fun “extras” such as the ability to save searches and execute “cross-database” searches

      How does each database handle controlled vocab when more than one database is searched?

    3. but only EBSCOhost allows searching by series title

      You can search by publication title in Proquest.

    4. Kentucky Virtual Library

      These are separate offerings: UKY and KVL. Correct me if I'm wrong, though.

    1. If this deep understanding is at the information seeker’s disposal, not having to set up advanced searches by applying drop down menus or clicking limiting boxes, and instead simply typing lines into a single command line search box could potentially be much more efficient.

      In fact, you could create a library of these searches!

    2. However, a cursory examination of the ProQuest results proved very few were actually relevant to my information needs concerning classical guitar in public schools, while six of the eight EBSCOhost results were highly useful.

      Search queries do not easily translate from one database to another because the contents of the databases, albeit similar, will determine what's returned. So this is a good example of having to revise the query and trying again.

    3. Of course, neither company is responsible for misinformation or quality contained within specific articles, but returned items from both databases include cross-references to authors, the subject headings applied to the content, and the full-text version of the work, when possible.

      This is a good point and a good clarification about what accuracy means with respect a discussion on databases.

    1. In considering these two particular databases, however, command line searching is only available as an explicitly framed option in IBSS.

      ah! Good (see comment above)

    2. Command line searching is available for both databases

      Do you mean advanced searching? EBSCOhost does allow for the construction of a command line syntax, but they don't offer a convenient place to write such a query. Instead you have to enter a command line query in one of the search boxes.

    3. Anthropology Plus, for instance, will produce search results for the truncated keywords “anthro*” AND “librar*” without comment. IBSS, by way of comparison, will produce a line of warning for the same search that reads: “Did you mean: anthro* AND library*?”

      I'm not following. Plus, wouldn't the suggestion help users?

    4. For instance, a search for the keywords

      In which of their databases?

    1. The subject headings are more detailed in EBSCO host than in ProQuest. In that regards, EBSCO host seems more accurate. For example, in an article on Northern Ireland,

      Might it vary based on the specific database that's being used?

    2. It feels like

      How would you examine that in order to know for sure?

    1.  No thesaurus is provided.

      An index is provided, though. How does that help?

    2. The text-only interface is complaint with Section 508 Standards.

      Nice observation---this is very important.

    1. If I were to narrow the search only focusing on certain publications, in certain languages, during a select period of time, the results would be more closely linked with what I was looking for. 

      The key here would be to focus more on faceting what we want to about global warming.

    2. EBSCOhost is compiled of all relevant databases

      What do these statements mean?

    1. s((http://adat.crl.edu)

      Nice use of outside source.

    2. In my opinion, EBSCOhost is more well recognized and respected by students, whereas ProQuest is not as well known.

      Why do you think that is?

    1. Many times with databases, it is to determine who should be published, and who is creditable enough to be published.

      I think 'indexed' might be a better choice. Publishers publish.

    1. Evaluating the accuracy of the database providers means considering the subject indexing.

      I think this is a spot on observation.

    2. Users can search within more than one database by check marking to select multiple titles

      This raises a good question: if a thesaurus is offered in one database and not another, and a user selects both databases, is the thesaurus still offered when the user conducts a search?

    1. are updated at varies times

      Do they tell you when for each one?

    1. Art Index enables a user to sort by: relevance; date newest; date oldest; source, and author. IBA allows sorting by: relevance, oldest first, and most recent first.

      given the structure of the records, could more sorting options be provided?

    2. (W)

      Must be a WordPress gotcha.

    3. Command Line

      I wonder how well command line search works from the phone!

    4. EBSCOhost provides a list of its databases in an alphabetical order. ProQuest Databases arranges its databases alphabetically in two forms, by name and by subject.

      What do you think about this? Does it matter?

    1. 127 databases in seven different subject areas

      Some of these repeat -- e.g., Alt-PressWatch is listed un the Arts databases and also the Language and Literature databases.

    2. full-text articles

      Indexed too -- full text will be a bit more limited. Would love to see some examples.