Author Archives: pgfeldman

Phil 3.1.18

7:00 – 4:30 ASRC MKT

Anonymize (done) and submit paper – done!
Finish T’s timeline approach? Finished my version. I think I like it.
This may be important: https://twitter.com/jack/status/969234275420655616
- We’re committing Twitter to help increase the collective health, openness, and civility of public conversation, and to hold ourselves publicly accountable towards progress.11:33 AM – 1 Mar 2018 from San Francisco, CA
  Our friends at @cortico and @socialmachines introduced us to the concept of measuring conversational health. They came up with four indicators: shared attention, shared reality, variety of opinion, and receptivity. Read about their work here: https://www.cortico.ai/blog/2018/2/29/public-sphere-health-indicators
- We simply can’t and don’t want to do this alone. So we’re seeking help by opening up an RFP process to cast the widest net possible for great ideas and implementations. This will take time, and we’re committed to providing all the necessary resources. RFP: https://blog.twitter.com/official/en_us/topics/company/2018/twitter-health-metrics-proposal-submission.html
Interactive topic hierarchy revision for exploring a collection of online conversations
- In the last decade, there has been an exponential growth of asynchronous online conversations (e.g. blogs), thanks to the rise of social media. Analyzing and gaining insights from such discussions can be quite challenging for a user, especially when the user deals with hundreds of comments that are scattered around multiple different conversations. A promising solution to this problem is to automatically mine the major topics from conversations and organize them into a hierarchical structure. However, the resultant topic hierarchy can be noisy and/or it may not match the user’s current information needs. To address this problem, we introduce a novel human-in-the-loop approach that allows the user to revise the topic hierarchy based on her feedback. We incorporate this approach within a visual text analytics system that helps users in analyzing and getting insights from conversations by exploring and revising the topic hierarchy. We evaluated the resulting system with real users in a lab-based study. The results from the user study, when compared to its counterpart that does not support interactive revisions of a hierarchical topic model, provide empirical evidence of the potential utility of our system in terms of both performance and subjective measures. Finally, we summarize generalizable lessons for introducing human-in-the-loop computation within a visual text analytics system
Understanding the Promise and Limits of Automated Fact-Checking
- The furor over so-called ‘fake news’ has exacerbated long-standing concerns about political lying and online rumors in a fragmented media environment, drawing attention to the potential of various automated fact-checking (AFC) technologies to combat online misinformation. This factsheet gives an overview of current efforts to automatically police false claims and misleading content online. Based on a review of recent research and interviews with both fact-checkers and computer scientists working in this area, we find that:
  - Much of the terrain covered by human fact-checkers requires a kind of judgement and sensitivity to context that remains far out of reach for fully automated verification.
  - Despite progress in automatic verification of a narrow range of simple factual claims, AFC systems will require human supervision for the foreseeable future.
  - The promise of AFC technologies for now lies in tools to assist fact-checkers to identify and investigate claims, and to deliver their conclusions, as effectively as possible.
More BIC
- Now it is the case, and increasingly widely recognized to be, that in games in general there’s no way players can rationally deliberate to a Nash equilibrium. Rather, classical canons of rationality do not in general support playing in Nash equilibria. So it looks as though shared intentions cannot, in the general run of games, by classical canons, be rationally formed! And that means in the general run of life as well. This is highly paradoxical if you think that rational people can have shared intentions. The paradox is not resolved by the thought that when they do, the context is not a game: any situation in which people have to make the sorts of decisions that issue in shared intentions must be a game, which is, after all, just a situation in which combinations of actions matter to the combining parties. (pg 139)
- Turn to the idea that a joint intention to do (x,y) is rationally produced in 1 and 2 by common knowledge of two conditional intentions: Pl has the intention expressed by ‘I’ll do x if and only if she does y’, and P2 the counterpart one. Clearly P1 doesn’t have the intention to do x if and. only if P2 in fact does y whether or not Pl believes P2 will do y; the right condition must be along the lines of:
  (C1) P1 intends to do x if and only if she believes P2 will do y. (pg 139)
  - So this is in belief space, and belief is based on awareness and trust
- There are two obstacles to showing this, one superable, the other not, I think. First, there are two Nash equilibria, and nothing in the setup to suggest that some standard refinement (strengthening) of the Nash equilibrium condition will eliminate one. However, I suspect that my description of the situation could be refined without ‘changing the subject’. Perhaps the conditional intention Cl should really be ‘I’ll do x if and only if she’ll do y, and that’s what I would like best’. For example, if x and y are the two obligations in a contract being discussed, it is natural to suppose that Pl thinks that both signing would be better than neither signing. If we accept this gloss then the payoff structure becomes a Stag Hunt – Hi-Lo if both are worse off out of equilibrium than in the poor equilibrium (x’ ,y’). To help the cause of rationally deriving the joint intention (x,y), assume the Hi-Lo case. What are the prospects now? As I have shown in chapter 1, there is no chance of deriving (x,y) by the classical canons, and the only (so far proposed) way of doing to is by team reasoning. (pg 140)
- The nature of team reasoning, and of the conditions under which it is likely to be primed in individual agents, has a consequence that gives further support to this claim. This is that joint intentions arrived at by the route of team reasoning involve, in the individual agents, a ‘sense of collectivity’. The nature of team reasoning has this effect, because the team reasoner asks herself not ‘What should I do?’ but ‘What should we do?’ So, to team-reason, you must already be in a frame in which first-person plural concepts are activated. The priming conditions for team reasoning have this effect because, as we shall see later in this chapter, team reasoning, for a shared objective, is likely to arise spontaneously in an individual who is in the psychological state of group-identifying with the set of interdependent actors; and to self-identify as a member of a group essentially involves a sense of collectivity. (pg 141)
Starting on ONR white paper – first pass banged together
- Need to add figures and references
discovered pandoc, which converts nicely between many files, including LaTex and word. The command that matters is:
```
pandoc -s foo.tex -o foo.docx
```

Phil 2.28.18

7:00 – 4:00 ASRC MKT

More BIC
- One of the things that MB seems to be saying here is that group identification has two parts. First is the self-identification with the group. Second is the mechanism that supports that framing. You can’t belong to a group you don’t see.
- To generalize the notions of team mechanism and team to unreliable contexts, we need the idea of the profile that gets enacted if all the agents function under a mechanism. Call this the protocol delivered by the mechanism. The protocol is , roughly, what everyone is supposed to do, what everyone does if the mechanism functions without any failure. But because there may well be failures, the protocol of a mechanism may not get enacted, some agents not playing their part but doing their default actions instead. For this reason the best protocol to have is not in general the first-best profile o*. In judging mechanisms we must take account of the states of the world in which there are failures, with their associated probabilities. How? Put it this way: if we are choosing a mechanism, we want one that delivers the protocol that maximizes the expected value of U. (pg 131)
- Group identification is a framing phenomenon. Among the many different dimensions of the frame of a decision-maker is the ‘unit of agency’ dimension: the framing agent may think of herself as an individual doer or as part of some collective doer. The first type of frame is operative in ordinary game-theoretic, individualistic reasoning, and the second in team reasoning. The concept-clusters of these two basic framings center round ‘I/ she/he’ concepts and ‘we’ concepts respectively. Players in the two types of frame begin their reasoning with the two basic conceptualizations of the situation, as a ‘What shall I do?’ problem, and a ‘What shall we do?’ problem, respectively. (pg 137)
Analyzing the Digital Traces of Political Manipulation: The 2016 Russian Interference Twitter Campaign
- Until recently, social media was seen to promote democratic discourse on social and political issues. However, this powerful communication platform has come under scrutiny for allowing hostile actors to exploit online discussions in an attempt to manipulate public opinion. A case in point is the ongoing U.S. Congress investigation of Russian interference in the 2016 U.S. election campaign, with Russia accused of, among other things, using trolls (malicious accounts created for the purpose of manipulation) and bots (automated accounts) to spread misinformation and politically biased information. In this study, we explore the effects of this manipulation campaign, taking a closer look at users who re-shared the posts produced on Twitter by the Russian troll accounts publicly disclosed by U.S. Congress investigation. We collected a dataset with over 43 million elections-related posts shared on Twitter between September 16 and October 21, 2016 by about 5.7 million distinct users. This dataset included accounts associated with the identified Russian trolls. We use label propagation to infer the ideology of all users based on the news sources they shared. This method enables us to classify a large number of users as liberal or conservative with precision and recall above 90%. Conservatives retweeted Russian trolls about 31 times more often than liberals and produced 36 times more tweets. Additionally, most retweets of troll content originated from two Southern states: Tennessee and Texas. Using state-of-the-art bot detection techniques, we estimated that about 4.9% and 6.2% of liberal and conservative users respectively were bots. Text analysis on the content shared by trolls reveals that they had a mostly conservative, pro-Trump agenda. Although an ideologically broad swath of Twitter users were exposed to Russian Trolls in the period leading up to the 2016 U.S. Presidential election, it was mainly conservatives who helped amplify their message.
CHIIR Talk
- Make new IR-Context graphic – done!
- De-uglify JuryRoom – done!
TensorFlow’s Machine Learning Crash Course

Phil 2.27.18

7:00 – 5:00 ASRC MKT

More BIC
- A mechanism is a general process. The idea (which I here leave only roughly stated) is of a causal process which determines (wholly or partly) what the agents do in any simple coordination context. It will be seen that all the examples I have mentioned are of this kind; contrast a mechanism that applies, say, only in two-person cases, or only to matching games, or only in business affairs. In particular, team reasoning is this kind of thing. It applies to any simple coordination context whatsoever. It is a mode of reasoning rather than an argument specific to a context. (pg 126)
- In particular, [if U is Paretian] the correct theory of Hi-Lo says that all play A. In short, an intuition in favour of C’ supports A-playing in Hi-Lo if we believe that all players are rational and there is one rationality. (pg 130)
  - Another form of dimension reduction – “We are all the same”
Machine Theory of Mind
- We design a Theory of Mind neural network – a ToMnet – which uses meta-learning to build models of the agents it encounters, from observations of their behaviour alone. Through this process, it acquires a strong prior model for agents’ behaviour, as well as the ability to bootstrap to richer predictions about agents’ characteristics and mental states using only a small number of behavioural observations. We apply the ToMnet to agents behaving in simple gridworld environments, showing that it learns to model random, algorithmic, and deep reinforcement learning agents from varied populations, and that it passes classic ToM tasks such as the “SallyAnne” test of recognising that others can hold false beliefs about the world
Classifier Technology and the Illusion of Progress (David Hand, 2006)
- A great many tools have been developed for supervised classification, ranging from early methods such as linear discriminant analysis through to modern developments such as neural networks and support vector machines. A large number of comparative studies have been conducted in attempts to establish the relative superiority of these methods. This paper argues that these comparisons often fail to take into account important aspects of real problems, so that the apparent superiority of more sophisticated methods may be something of an illusion. In particular, simple methods typically yield performance almost as good as more sophisticated methods, to the extent that the difference in performance may be swamped by other sources of uncertainty that generally are not considered in the classical supervised classification paradigm.
Sensitivity and Generalization in Neural Networks: an Empirical Study
- Neural nets generalize better when they’re larger and less sensitive to their inputs, are less sensitive near training data than away from it, and other results from massive experiments. (From @Jascha)
- The graph represents a network of 6,716 Twitter users whose recent tweets contained “#NIPS2017”, or who were replied to or mentioned in those tweets, taken from a data set limited to a maximum of 18,000 tweets. The network was obtained from Twitter on Friday, 08 December 2017 at 15:30 UTC.
Back to Basics: Benchmarking Canonical Evolution Strategies for Playing Atari
- Evolution Strategies (ES) have recently been demonstrated to be a viable alternative to reinforcement learning (RL) algorithms on a set of challenging deep RL problems, including Atari games and MuJoCo humanoid locomotion benchmarks. While the ES algorithms in that work belonged to the specialized class of natural evolution strategies (which resemble approximate gradient RL algorithms, such as REINFORCE), we demonstrate that even a very basic canonical ES algorithm can achieve the same or even better performance. This success of a basic ES algorithm suggests that the state-of-the-art can be advanced further by integrating the many advances made in the field of ES in the last decades.
  We also demonstrate qualitatively that ES algorithms have very different performance characteristics than traditional RL algorithms: on some games, they learn to exploit the environment and perform much better while on others they can get stuck in suboptimal local minima. Combining their strengths with those of traditional RL algorithms is therefore likely to lead to new advances in the state of the art.
Copied over SheetToMap to the Applications file on TOSHIBA
Created a Data folder, which has all the input and output files for the various applications
Need to add a curDir variable to LMN
Presentation:
- I need to put together a 2×2 payoff matrix that covers nomad/flock/stampede – done
- Some more heat map views, showing nomad, flocking – done
- De-uglify JuryRoom
- Timeline of references – done
- Collapse a few pages 22.5 minutes for presentation and questions – done
Start on white paper

Phil 2.26.18

7:00 – 6:00 ASRC MKT

Spread of information is dominated by search ranking
- Twitter thread
  - The spreading process was linear because the background search rate is roughly constant day to day for discounts, and any viral element turned out to be quite small.
- Paper
BIC
- There are many conceivable team mechanisms apart from simple direction and team reasoning; they differ in the way in which computation is distributed and the pattern of message sending. For example, one agent might compute o* and send instructions to the others. With the exception of team reasoning, these mechanisms involve the communication of information. If they do I shall call them modes of organization or protocols. (pg 125)
- A mechanism is a general process. The idea (which I here leave only roughly stated) is of a causal process which determines (wholly or partly) what the agents do in any simple coordination context. It will be seen that all the examples I have mentioned are of this kind; contrast a mechanism that applies, say, only in two-person cases, or only to matching games, or only in business affairs. In particular, team reasoning is this kind of thing. It applies to any simple coordination context whatsoever. It is a mode of reasoning rather than an argument specific to a context. (pg 126)
Presentation:
- I need to put together a 2×2 payoff matrix that covers nomad/flock/stampede
- Some more heat map views, showing nomad, flocking
- De-uglify JuryRoom
- Timeline of references
- Collapse a few pages 22.5 minutes for presentation and questions
Work on getting SheetToMap in a swing app? Less figuring things out…
- Slower going than I hoped, but mostly working now. As always, StackOverflow to the rescue: How to draw graph inside swing with GraphStream actually?
- Adding load and save menu choices. Done! Had a few issues with getting the position of the nodes saved out. It seems like you should do this?
```
GraphicNode gn = viewer.getGraphicGraph().getNode(name);
row.createCell(cellIndex++).setCellValue(gn.getX());
row.createCell(cellIndex++).setCellValue(gn.getY());
```
- Anyway, pretty pix:
Start on white paper
Fika

Phil 2.25.18

Looks like I need to update the DC and the CI 2018 paper with a new reference:

Dynamic Word Embeddings for Evolving Semantic Discovery

Zijun Yao, Yifan Sun, Weicong Ding, Nikhil Rao, Hui Xiong
Word evolution refers to the changing meanings and associations of words throughout time, as a byproduct of human language evolution. By studying word evolution, we can infer social trends and language constructs over different periods of human history. However, traditional techniques such as word representation learning do not adequately capture the evolving language structure and vocabulary. In this paper, we develop a dynamic statistical model to learn time-aware word vector representation. We propose a model that simultaneously learns time-aware embeddings and solves the resulting “alignment problem”. This model is trained on a crawled NYTimes dataset. Additionally, we develop multiple intuitive evaluation strategies of temporal word embeddings. Our qualitative and quantitative tests indicate that our method not only reliably captures this evolution over time, but also consistently outperforms state-of-the-art temporal embedding approaches on both semantic accuracy and alignment quality.

Phil 2.23.18

6:30 – 8:30, 11:00 – 5:00 ASRC MKT

Graphstream with javafx? https://github.com/graphstream/gs-ui-javafx
Learning to Cooperate, Compete, and Communicate
- Multiagent environments where agents compete for resources are stepping stones on the path to AGI. Multiagent environments have two useful properties: first, there is a natural curriculum — the difficulty of the environment is determined by the skill of your competitors (and if you’re competing against clones of yourself, the environment exactly matches your skill level). Second, a multiagent environment has no stable equilibrium: no matter how smart an agent is, there’s always pressure to get smarter. These environments have a very different feel from traditional environments, and it’ll take a lot more research before we become good at them.
Storytelling and Politics: How History, Myths and Narratives Drive Our Decisions (video)
- A narrative with historical overtones, an emotive connection and credibility not only convinces people, it frames the points of reference they use to evaluate the decision they are being asked to make.
- Logos Pathos Ethos?
Continuing with rewrite. Had to fire up the MiKTex admin console to install wrapfig. Permissions issue?
- Need to take the description of the maps at the end of the results section and turn into a paragraph.
Walk through of presentation this afternoon. Need to set up a skype session and bridge. Went well, I need to make a few fixes. Most importantly I need to put together a 2×2 payoff matrix that covers nomad/flock/stampede

Phil 2.22.18

7:00 – ASRC MKT

Long chat with Wajant about the CI 2018 paper. going to work up a new version
- Started in Docs, but wound up saving out and reworking the LaTex version to keep track of the length.
Coincidentally, ONR is soliciting white papers for theoretically-based decision making tools. Five pages plus references for the paper, and a one-page resume.
- The 5-page body of the white paper shall include the following information:
  - Principal Investigator(s);
  - Relevance of the proposed effort to the research areas described in Section II; (Topic 2, Research Focus Area 1)
    - relationship of the proposed work to current state of art.
  - Technical objective of the proposed effort;
  - Technical approach that will be pursued to meet the objective;
  - A summary of recent relevant technical breakthroughs; and
  - A funding plan showing requested funding per fiscal year.
Need to register for TF conference when Aaron gets in. Got hotel and $$ approval.
More dimension reduction and belief vectors on twitter

Phil 2.21.18

7:00 – 6:00 ASRC MKT

Wow – I’m going to the Tensorflow Summit! Need to get a hotel.
Dimension reduction + velocity in this thread
Global Pose Estimation with an Attention-based Recurrent Network
- The ability for an agent to localize itself within an environment is crucial for many real-world applications. For unknown environments, Simultaneous Localization and Mapping (SLAM) enables incremental and concurrent building of and localizing within a map. We present a new, differentiable architecture, Neural Graph Optimizer, progressing towards a complete neural network solution for SLAM by designing a system composed of a local pose estimation model, a novel pose selection module, and a novel graph optimization process. The entire architecture is trained in an end-to-end fashion, enabling the network to automatically learn domain-specific features relevant to the visual odometry and avoid the involved process of feature engineering. We demonstrate the effectiveness of our system on a simulated 2D maze and the 3D ViZ-Doom environment.
Slides
- Location
- Orientation
- Velocity
- IR context -> Sociocultural context
Writing Fika. Make a few printouts of the abstract
- It kinda happened. W
Write up LMN4A2P thoughts. Took the following and put them in a LMN4A2P roadmap document in Google Docs
- Storing a corpora (raw text, BoW, TF-IDF, Matrix)
  - Uploading from file
  - Uploading from link/crawl
  - Corpora labeling and exploring
- Index with ElasticSearch
- Production of word vectors or ‘effigy documents’
- Effigy search using Google CSE for public documents that are similar
  - General
  - Site-specific
  - Semantic (Academic, etc)
- Search page
  - Lists (reweightable) or terms and documents
  - Cluster-based map (pan/zoom/search)
I’m as enthusiastic about the future of AI as (almost) anyone, but I would estimate I’ve created 1000X more value from careful manual analysis of a few high quality data sets than I have from all the fancy ML models I’ve trained combined. (Thread by Sean Taylor on Twitter, 8:33 Feb 19, 2018)
Prophet is a procedure for forecasting time series data. It is based on an additive model where non-linear trends are fit with yearly and weekly seasonality, plus holidays. It works best with daily periodicity data with at least one year of historical data. Prophet is robust to missing data, shifts in the trend, and large outliers.

Done with Angular fundamentals. reDirectTo isn’t working though…

zone.js:405 Unhandled Promise rejection: Invalid configuration of route '': redirectTo and component cannot be used together ; Zone: <root> ; Task: Promise.then ; Value: Error: Invalid configuration of route '': redirectTo and component cannot be used together

Phil 2.20.18

7:00 – 5:00 ASRC MKT

Diversity injection: How to Inoculate the Public Against Fake News
- Cambridge researchers developed a game to help people understand, broadly, how fake news works by having users play trolls and create misinformation. By “placing news consumers in the shoes of (fake) news producers, they are not merely exposed to small portions of misinformation,” the researchers write in their accompanying paper.
Physics of human cooperation: experimental evidence and theoretical models
- Angel Sánchez (Scholar)
- In recent years, many physicists have used evolutionary game theory combined with a complex systems perspective in an attempt to understand social phenomena and challenges. Prominent among such phenomena is the issue of the emergence and sustainability of cooperation in a networked world of selfish or self-focused individuals. The vast majority of research done by physicists on these questions is theoretical, and is almost always posed in terms of agent-based models. Unfortunately, more often than not such models ignore a number of facts that are well established experimentally, and are thus rendered irrelevant to actual social applications. I here summarize some of the facts that any realistic model should incorporate and take into account, discuss important aspects underlying the relation between theory and experiments, and discuss future directions for research based on the available experimental knowledge.
What We Read, What We Search: Media Attention and Public Attention Among 193 Countries
- We investigate the alignment of international attention of news media organizations within 193 countries with the expressed international interests of the public within those same countries from March 7, 2016 to April 14, 2017. We collect fourteen months of longitudinal data of online news from Unfiltered News and web search volume data from Google Trends and build a multiplex network of media attention and public attention in order to study its structural and dynamic properties. Structurally, the media attention and the public attention are both similar and different depending on the resolution of the analysis. For example, we find that 63.2% of the country-specific media and the public pay attention to different countries, but local attention flow patterns, which are measured by network motifs, are very similar. We also show that there are strong regional similarities with both media and public attention that is only disrupted by significantly major worldwide incidents (e.g., Brexit). Using Granger causality, we show that there are a substantial number of countries where media attention and public attention are dissimilar by topical interest. Our findings show that the media and public attention toward specific countries are often at odds, indicating that the public within these countries may be ignoring their country-specific news outlets and seeking other online sources to address their media needs and desires.
Sent Jen a note about carpooling to CHIIR. Need to check out one day earlier
Add slides
- Two phases – theoretical model building, then study
- Implications for design based on
- Something about velocity? Academic journal papers (slow production, slow consumption) at one end and twitter on the other (fast production, fast consumption)
Ingesting Documents (pdf, word, txt, etc) Into ElasticSearch
More Angular
Discussions with Aaron about getting some LMN capability into A2P.

Phil 2.19.18

7:30 – 4:30 ASRC MKT

Back to BIC.
- (page 102)
- (pg 107)
- (pg 107)
- Sociality: Coordinating bodies, minds and groups
  - Human interaction, as opposed to aggregation, occurs in face-to-face groups. “Sociality theory” proposes that such groups have a nested, hierarchical structure, consisting of a few basic variations, or “core configurations.” These function in the coordination of human behavior, and are repeatedly assembled, generation to generation, in human ontogeny, and in daily life. If face-to-face groups are “the mind’s natural environment,” then we should expect human mental systems to correlate with core configurations. Features of groups that recur across generations could provide a descriptive paradigm for testable and non-intuitive evolutionary hypotheses about social and cognitive processes. This target article sketches three major topics in sociality theory, roughly corresponding to the interests of biologists, psychologists, and social scientists. These are (1) a multiple levels-of-selection view of Darwinism, part group selectionism, part developmental systems theory; (2) structural and psychological features of repeatedly assembled, concretely situated face-to-face coordination; and (3) superordinate, “unsituated” coordination at the level of large-scale societies. Sociality theory predicts a tension, perhaps unresolvable, between the social construction of knowledge, which facilitates coordination within groups, and the negotiation of the habitat, which requires some correspondence with contingencies in specific situations. This tension is relevant to ongoing debates about scientific realism, constructivism, and relativism in the philosophy and sociology of knowledge.
    - These definitions seem to span atomic (mother/child, etc), small group (situated, environmental), and societal (unsituated, normative)
  - Coordination occurs to the extent that knowledge and practice domains overlap or are complementary. I suggest that values serve as a medium. Humans live in a value-saturated environment; values are known from interactions with people, natural objects, and artifacts
    - Dimension reduction
I’m starting to think that agents as gradient descent machines within networks is something to look for:
- Individual Strategy Update and Emergence of Cooperation in Social Networks
  - In this article, we critically study whether social networks can explain the emergence of cooperative behavior. We carry out an extensive simulation program in which we study the most representative social dilemmas. For the Prisoner’s Dilemma, it turns out that the emergence of cooperation is dependent on the microdynamics. On the other hand, network clustering mostly facilitates global cooperation in the Stag Hunt game, whereas degree heterogeneity promotes cooperation in Snowdrift dilemmas. Thus, social networks do not promote cooperation in general, because the macro-outcome is not robust under change of dynamics. Therefore, having specific applications of interest in mind is crucial to include the appropriate microdetails in a good model.
- Alex Peysakhovich and Adam Lerer
  - Prosocial learning agents solve generalized Stag Hunts better than selfish ones
    - Deep reinforcement learning has become an important paradigm for constructing agents that can enter complex multi-agent situations and improve their policies through experience. One commonly used technique is reactive training – applying standard RL methods while treating other agents as a part of the learner’s environment. It is known that in general-sum games reactive training can lead groups of agents to converge to inefficient outcomes. We focus on one such class of environments: Stag Hunt games. Here agents either choose a risky cooperative policy (which leads to high payoffs if both choose it but low payoffs to an agent who attempts it alone) or a safe one (which leads to a safe payoff no matter what). We ask how we can change the learning rule of a single agent to improve its outcomes in Stag Hunts that include other reactive learners. We extend existing work on reward-shaping in multi-agent reinforcement learning and show that that making a single agent prosocial, that is, making them care about the rewards of their partners can increase the probability that groups converge to good outcomes. Thus, even if we control a single agent in a group making that agent prosocial can increase our agent’s long-run payoff. We show experimentally that this result carries over to a variety of more complex environments with Stag Hunt-like dynamics including ones where agents must learn from raw input pixels.
  - The Good, the Bad, and the Unflinchingly Selfish: Cooperative Decision-Making Can Be Predicted with High Accuracy Using Only Three Behavioral Types
    - The human willingness to pay costs to benefit anonymous others is often explained by social preferences: rather than only valuing their own material payoff, people also care in some fashion about the outcomes of others. But how successful is this concept of outcome-based social preferences for actually predicting out-of-sample behavior? We investigate this question by having 1067 human subjects each make 20 cooperation decisions, and using machine learning to predict their last 5 choices based on their first 15. We find that decisions can be predicted with high accuracy by models that include outcome-based features and allow for heterogeneity across individuals in baseline cooperativeness and the weights placed on the outcome-based features (AUC=0.89). It is not necessary, however, to have a fully heterogeneous model — excellent predictive power (AUC=0.88) is achieved by a model that allows three different sets of baseline cooperativeness and feature weights (i.e. three behavioral types), defined based on the participant’s cooperation frequency in the 15 training trials: those who cooperated at least half the time, those who cooperated less than half the time, and those who never cooperated. Finally, we provide evidence that this inclination to cooperate cannot be well proxied by other personality/morality survey measures or demographics, and thus is a natural kind (or “cooperative phenotype”)
    - “least”, “intermediate” and “most” cooperative. Doesn’t give percentages, though it says that 17.8% were cooperative?
Talk Susan Gregurick (susan.gregurick@nih.gov)
- All of Us research program
- Opiod epidemic – trajectory modeling?
- PZM21 computational drug
- Develop advanced software and tools. Specialized generalizable and accessible tools for biomedicing (finding stream). Includes mobile, data indexing, etc.
- NIH Data Fellows? Postdocs to senior industry
- T32 funding? Mike Summers at UMBC
- ncbi-hackathons.github.io (look for data?
- Primary supporter for machine learning is NIMH (imaging), then NIGNS, and NCI Team science (Multi-PI) is a developing thing
- $400m in computing enabled interactions (human in the loop decision tools. Research Browser?
- Big Data to Knowledge Initiative (BD2K) datascience.nih.gov/bd2k
- Interagency Modeling and Analysis Group (IMAG) imagewiki,nibib.nih.gov
- funding: bisti.nih.gov
- NIH RePorter projectreporter.nih.gov Check out matchmaker. What’s the ranking algorithm?
- NIDDK predictive analytics for budgeting <- A2P-ish?
- Most of thi srequires preliminary data and papers to be considered for funding. There is one opportunity for getting funding to get preliminary data. Need to get more specific infor here.
- Each SRO normalizes grade as a percentile, not the score, since some places inflate, and others are hard.
- Richard Aargon at NIGMS
- Office of behavioral and social science – NIH center Francis Collins. Also agent-based simulation
- Really wants a Research Browser to go through proposals
Fika – study design
- IRB – you can email and chat with the board if you have a tricky study

Phil 2.17.18

Snow today

Random thought. Marches help because they literally put people in the same position, align them in a direction and enforce a common velocity.

Found this item about compromised Trump company servers. I think this could be framed as an example of excessive trust bringing high social inertia. This would be different from explicit command and control.

So why would the organization allow this? If it’s not fear, then it’s trust. Kramer describes trust arising from incremental and repeated exchanges. I’d like to extend that thought. These incremental and repeated exchanges need to happen in dimension-reduced spaces that are similar or appear to map to each other, and occur with respect to orientation alignment, position, and velocity. The more alignment in these three axis, the greater the trust, and the lower the perceived need for awareness outside the relationship.

Twitter is just full of this stuff today: How A Russian Troll Fooled America

Reconstructing the life of a covert Kremlin influence account (TEN_GOP)
The Atlantic Council’s Digital Forensic Research Lab (DFRLab) has operationalized the study of disinformation by exposing falsehoods and fake news, documenting human rights abuses, and building digital resilience worldwide.

Phil 2.16.18

7:00 – 3:00 ASRC MKT

Finished the first draft of the CI 2018 extended abstract!
And I also figured out how to run the sub projects in the Ultimate Angular src collection. You need to go to the root directory for the chapter, run yarn install, then yarn start. Everything works then.
Trolls on Twitter: How Mainstream and Local News Outlets Were Used to Drive a Polarized News Agenda
- This is the kind of data that compels us to rethink how we understand Twitter — and what I feel are more influential platforms for reaching regular people that include Facebook, Instagram, Google, and Tumblr, as well as understand ad tech tracking and RSS feed–harvesting as part of the greater propaganda ecosystem.
NELA News credibility classification toolkit
- The News Landscape (NELA) Toolkit is an open source toolkit for the systematic exploration of the news landscape. The goal of NELA is to both speed up human fact-checking efforts and increase the understanding of online news as a whole. NELA is made up of multiple indepedent modules, that work at article level granularity: reliability prediction, political impartiality prediction, text objectivity prediction, and reddit community interest prediction. As well as, modules that work at source level granularity: reliability prediction, political impartiality prediction, content-based feature visualization.
New benchmarks for approximate nearest neighbors
- I built ANN-benchmarksto address this. It pits a bunch of implementations (including Annoy) against each other in a death match: which one can return the most accurate nearest neighbors in the fastest time possible. It’s not a new project, but I haven’t actively worked on it for a while.
Systems of Global Governance in the Era of Human-Machine Convergence
- Technology is increasingly shaping our social structures and is becoming a driving force in altering human biology. Besides, human activities already proved to have a significant impact on the Earth system which in turn generates complex feedback loops between social and ecological systems. Furthermore, since our species evolved relatively fast from small groups of hunter-gatherers to large and technology-intensive urban agglomerations, it is not a surprise that the major institutions of human society are no longer fit to cope with the present complexity. In this note we draw foundational parallelisms between neurophysiological systems and ICT-enabled social systems, discussing how frameworks rooted in biology and physics could provide heuristic value in the design of evolutionary systems relevant to politics and economics. In this regard we highlight how the governance of emerging technology (i.e. nanotechnology, biotechnology, information technology, and cognitive science), and the one of climate change both presently confront us with a number of connected challenges. In particular: historically high level of inequality; the co-existence of growing multipolar cultural systems in an unprecedentedly connected world; the unlikely reaching of the institutional agreements required to deviate abnormal trajectories of development. We argue that wise general solutions to such interrelated issues should embed the deep understanding of how to elicit mutual incentives in the socio-economic subsystems of Earth system in order to jointly concur to a global utility function (e.g. avoiding the reach of planetary boundaries and widespread social unrest). We leave some open questions on how techno-social systems can effectively learn and adapt with respect to our understanding of geopolitical complexity.

Phil 2.15.18

ASRC MKT 7:00 – 8:00

Taking most of the day off, but spent the early morning tweaking the CI 2018 paper and sending it out to the Fika writing group
We have discussions, but we do not have discussions about the axis that we are choosing to decide along

Sent this to my representative:

Dear Rep. Cummings,

I would like to suggest a simple piece of legislation that may begin to address gun violence.

“For every student killed or wounded with a firearm in the preceding year, a 1-cent tax will be added to the price of the type of bullet used in the attack. The funds collected will be used to support the victims.”

This approach will do two things: 1) It will incentivize gun owners to demand action, since it could substantially increase the cost of using their guns. 2) It will place the onus of determining effective gun control within the gun community. As a result, there should be no second amendment concerns.

I realize that this is small in scope, and targeted at only the most innocent victims of gun violence, but I’m hoping that the simplicity and strength of the message may help moving the process forward.

Phil 2.14.18

7:00 – 4:00 ASRC

Stampede? Herding? Twitter deleted 200,000 Russian troll tweets. Read them here.
- Twitter doesn’t make it easy to track Russian propaganda efforts — this database can help
Add a “show all trajectories” checkbox.
- That’s a nice visualization that shows the idea of the terrain uncovered by the trajectories:
Continue with paper – down to 3 pages!
Continue with slides. Initial walkthrough with Aaron
3:00 – 4:00 A2P meeting

Phil 2.13.18