This is a list of all the questions and their associated study carrel identifiers. One can learn a lot of the "aboutness" of a text simply by reading the questions.
| identifier | question |
|---|---|
| jiang | So how can we start and manage successful cross- disciplinary research? |
| jiang | What can we do to facilitate collaborative behaviors? |
| cohen-nakazawa | But how should one handle the lack of a specific vocabulary within a given training set? |
| cohen-nakazawa | How, we asked, should one handle such instances of effacement? |
| harper | Comment by Charlie Harper: Does this better explain word generation with a Markov model? |
| harper | What is Generative Machine Learning? |
| harper | Z Generator Discriminator Real or Fake? |
| harper | “ Can Adversarial Networks Hallucinate Occluded People With a Plausible Aspect? ” Computer Vision and Image Understanding 182( May): 71- 80. |
| morgan | For example, who predicted the demise of many centralized information authorities? |
| morgan | How many librarians are able to create a file, put it on the Web, and share the resulting URL? |
| morgan | Yet again, libraries use the Internet, but do they actually exploit it? |
| lucic-shanahan | How did different neighborhoods in a diverse but also segregated city respond to particular book choices? |
| lucic-shanahan | How do libraries innovate? |
| lucic-shanahan | What if you combined the last two paragraphs with some additional glue, as follows? |
| hansen | Can a Hammer Categorize Highly Technical Articles? |
| hansen | Comment by Mark Dehmlow: Is this meant to be MARC subfields or professional subfields? |
| hansen | Comment by Mark Dehmlow: Maybe pursued? |
| hansen | Comment by Mark Dehmlow: Maybe use"the project?? |
| hansen | Comment by Mark Dehmlow: Maybe use"the project?? |
| hansen | Comment by Mark Dehmlow: Maybe"approach"? |
| hansen | Comment by Mark Dehmlow: Maybe"think analytically"? |
| hansen | Comment by Mark Dehmlow: Perhaps a different word choice? |
| hansen | Is machine learning an action or a technique? |
| hansen | Maybe something like problem instead of project OR supporting instead of solving? |
| hansen | Not to mention: is this really a problem or are we just looking for a way to employ machine learning to say that we did? |
| hansen | Perhaps a short footnote? |
| hansen | Questions like: is this problem large enough in scale to substitute machines for human labor given that machines will likely be less accurate? |
| altman | Can you find examples of success( or failure) from people using similar datasets with the same algorithm? |
| altman | Can you prepare your data with a reasonable amount of pre- processing? |
| altman | Comment by Daniel Johnson: Am I parsing this right? |
| altman | Equally important, what would you expect to see if the model did n’t learn anything at all? |
| altman | Is it important that your algorithm be easy to explain? |
| altman | Start by asking: What would you expect to see if the model learned perfectly? |
| altman | · Are the tools stable, well- documented, and well- supported? |
| altman | · Are you concerned about speed and scale? |
| altman | · Can you explain how this algorithm works to a lay person and why it ’s beneficial to the project? |
| altman | · How do the tools align with the skill sets and comfort levels of your team? |
| altman | · How well do tools interoperate with one another, or with other parts of your existing systems? |
| altman | · What do the inputs and outputs of the algorithm mean? |
| altman | · What effect do different hyperparameters have on the algorithm? |
| altman | · What sorts of features and interfaces do they offer? |
| hintze-schossau | Are these quantitative methods sneaking into the humanities disguised by anthropomorphic terms like a wolf shrouded in a sheep ’s fleece? |
| hintze-schossau | Are they capable of developing their own perspective, and can they aid in qualitative research? |
| hintze-schossau | Artificial Intelligence in the Humanities: Wolf in Disguise, or Digital Revolution? |
| hintze-schossau | But are these technologies imbued with intuition or understanding, and do they learn like humans? |
| hintze-schossau | Did you read a manual before you used your smartphone, or did you like everyone else just “ figure it out ”? |
| hintze-schossau | Do we speak for everyone? |
| hintze-schossau | How do we understand what is essential? |
| hintze-schossau | Is it an intelligent machine, or a human trapped in a machine? |
| hintze-schossau | Is sunshine better than rain? |
| hintze-schossau | Now that we know the facts, how should we choose? |
| hintze-schossau | So, if these important sections in our imaginary machine learning book do not discuss the mechanisms of learning, then what are they discussing? |
| hintze-schossau | Trained Machines: Tool or Provocation? |
| hintze-schossau | What are our values? |
| hintze-schossau | “ What Is Memory? |
| prudhomme | Are Data Ready? ” NLM Musings from the Mezzanine( blog). |
| prudhomme | Comment by Mark Dehmlow: Maybe use active voice? |
| prudhomme | For images, how can archives build a data- labeling pipeline into their digital curation workflow that enables machine learning of collections? |
| prudhomme | How can cultural heritage institutions ensure that their machine learning algorithms avoid such bad outputs? |
| prudhomme | How much could we rely on the established institutional memory within the archives? |
| prudhomme | In short, how could we justify the value of these digital assets for curatorial purposes? |
| prudhomme | Lastly, could machine learning or deep learning applications help us build a greater capacity to augment knowledge? |
| prudhomme | What started this Pilot Project? |
| prudhomme | Why do they Matter? |
| prudhomme | “ What is Artificial Intelligence? ” Professor John McCarthy( website). |
| prudhomme | “ What is Keras? |
| lesk | But if software has been trained on a small and specific subset, will it generalize? |
| lesk | But, what if I leave off the final period when I enter it in Google Translate? |
| lesk | Can we rely on such software? |
| lesk | If image recognition systems can be fooled by slight changes in an image, what happens if self- driving cars misread road signs? |
| lesk | Recognition of problems Can we tell when something is wrong? |
| lesk | Somebody who knew statistics perked up and said “ it ’s a 50% chance of recognizing the animal? |
| lesk | Sometimes you have to use categorical variables( what county do you live in?) |
| lesk | What did we do before we had supercomputers? |
| lesk | What should we do? |
| lesk | What would I suggest? |
| lesk | Why not extrapolate? |
| lesk | Will face recognition software trained on adult white male faces work with other races, with women, and with children? |
| lesk | “ Is Artificial Intelligence Permanently Inscrutable? ” Nautilus 40( 1). |
| wiegand | Can the library work with other cultural heritage institutions to highlight the unique contributions of all? |
| wiegand | Discussion The larger questions now are, who is heard and who contributes? |
| wiegand | How are gaps, identified in needs analysis, reduced? |
| wiegand | How best to communicate with stakeholders about the value of the distinctive library mission? |
| wiegand | How can librarians best teach skills, foster engagement, and create knowledge to make a distinctive contribution to the institution? |
| wiegand | How can librarians reconcile past identity, add value, and leverage hard- won expertise in a new environment? |
| wiegand | Libraries could start with analyses of needs — what problems could be solved with more effective literature searches? |
| wiegand | What are sources of funding for libraries to develop this important work and not leave it to commercial services? |
| wiegand | What does the library do that no other entity on campus can do? |
| wiegand | What else might a researcher want to do that the library could include in a Researchers ’ Workstation? |
| wiegand | What is a library in rapidly changing times? |
| wiegand | What kind of teaching could help build citizens and critical thinkers, rather than simply expanding consumers of content? |
| wiegand | What research could fill gaps and inform solutions to those needs? |
| kim | AI Applications for Libraries Do these ethical concerns outlined above apply to libraries? |
| kim | Between the two moral rules, “ do not kill a person ” and “ save lives, ” which one should trump the other? |
| kim | But how far should one go to maximize the good? |
| kim | But how would an AI system be able to do so, and should they be allowed to? |
| kim | But what does “ as intelligent as a human ” exactly mean? |
| kim | But what if things were different and we were comparing five school children versus one baby or five babies versus one schoolchild? |
| kim | But what if those moral rules conflict with one another? |
| kim | Does the AI- powered information environment have a similar problem? |
| kim | In such a case, can we be justified in letting robots take the lives of other human beings? |
| kim | Is it permissible for people to delegate such a decision to robots? |
| kim | Should one program the AI- powered trolley to swerve or stay on course? |
| kim | Should you pull the lever, so that the runaway trolley would kill one person instead of five? |
| kim | What if the five people on the track are helpless babies or murderers who just escaped from the prison? |
| kim | What kind of AI- driven information systems and environments would be created if libraries approach AI with such intention from the beginning? |
| kim | Would or should that affect our decision? |
| kim | Would this count as a right thing to do? |
| kim | “ How Many Computers to Identify a Cat? |
| kim | “ What Are Jus Ad Bellum and Jus in Bello?, ” September. |
| kim | “ Who Decides: Man or Machine? ” Armed Forces Journal, November. |