Category Archives: Conferences

Phil 12.7.18

7:00 – 4:30 ASRC NASA/PhD

Phil 10.15.18

7:00 – ASRC BD

  • Heard about some interesting things this morning on BBC Business Daily – Is the Internet Fit for Purpose?:
    • Future in Review Conference: The leading global conference on the intersection of technology and the economy. New partnerships, projects, and plans you can’t afford to miss. If your success depends on having an accurate view of the future, or you’d like to meet others who are able and motivated to forge action-based alliances, this is the most important conference you will attend. Be one of the thought leaders in the FiRe conversation, analyzing and creating the future of technology, economics, pure science, the environment, genomics, education, and more.
    • Berit Anderson. Created the science fact/fiction magazine Scout, which, interestingly enough, has a discussion space for JuryRoom-style questions
  • More DARPA proposal

Phil 10.4.18

7:00 – 5:30 ASRC MKT

  • Join PCA! Write classified! Done
  • There are 56 work days until Jan 1. My 400 hours is 50 days. So I go full time on research around the 22nd.
  • Got a note from Wayne saying that there were 25 blue sky papers and 3 slots. THat might me expanded to 6 slots
  • Write up notes on “At Home in the Universe” – started
  • Finish speaking notes for BAA – Done
  • Matt found a couple of things that might be good. One is due on October 16th, which is waaaaaaaaaaaaayyyyyyyy too tight.
  • Looked at the Connected Health clearinghouse effort and website. It sounds a lot like a military version of PubMed, with the ability to request reports on demand, plus some standardized reports as well. These reports seem to source back to other agencies like the CDC, with external SMEs.

Phil 10.3.18

7:00 – 5:30 ASRC MKT

  • Finished At Home in the Universe. Really good. I’ll work on writing up notes this evening. The Kindle clippings feature is awesome
  • The stampeding robots paper is up on ArXiv: Disrupting the Coming Robot Stampedes: Designing Resilient Information Ecologies
  • Dopamine modulates novelty seeking behavior during decision making.
  • Need to finish Antonio’s paper, but my sense at this point is to add our work as a discussion of edge conditions that come up in the discussion section?
    • Done. Sent a letter discussing NIST RCS
  • Need to write up the fitness landscape thoughts. One axis is distance to model which is has a decay radius from each agent. Another axis is the price of an item(with future discounting?). Another axis is cost by agent to acquire the item. Cluster behavior emerges from local agents trying to find the best model and acquire the most value? There is also some kind of explicit connection between individuals that needs to be handled (a tanker and a plane have a client-server relationship that requires them to move in a coordinated way)
    • There is also information that is within the agents, and information that is in the environment. There may be other types of information as well.
  • Get Matt rolling on the whitepaper? – done!
  • Watson backend to A2P?
  • Kibitzed Aaron on how to access style sheets
  • Got about halfway through speaking notes on Army BAA

Phil 9.25.18

7:00 – 5:00 ASRC MKT

  • Wayne’s notes from yesterday:
    • Part of the wrapper for this will be why these issues might matter for the iSchool’s research future. I can help with the framing there.
      Yikes, 4 pages in this format? That is nothing!
      Will really have to shave this down to the absolute minimum.
      To that end I think the scenarios get fleshed out in their fullest now to capture all of the ideas and then hacked brutally into 1-2 paragraphs.
      The abstract probably goes to 4 sentences.
      Images stay, but no larger.
      We’ll work this out, but, man, that is barely 1500 words. Who was thinking when they put this together? 😉
  • Want to redo the designed system chart so that the complexity zone is concave – done.
  • More writing. Figured out that cars would be crashing at a rate of 3-4/sec based on 2016 data. Yikes!
  • Worked with Aaron on response to Antonio’s proposal. IEEE Software is a “production” magazine. And a nice marker for production is what kind of libraries are available, because then articles can be written on how to use them.
  • Kate Starbird this Friday! 10:00am – 12:00pm 2119 Hornbake Library South Wing
  • There is a world nomad games

Phil 7.9.18

SASO 2018

Trust in Organizations: Frontiers of Theory and Research

  • Although its importance is readily apparent, the contours of trust in collective contexts are much less obvious. The decision to trust in collective settings is different from, and in many respects more problematic than, decisions about trust that arise in other social contexts. Because of the size and structural complexity of large organizations, for example, individuals do not have the opportunity to engage in the sort of incremental and repeated exchanges that have been shown to facilitate the development of trust in more intimate settings, such as dyadic relationships

Keynote 4 – Heiko Hamann (chair R.P. Würtz, room: FBK – Sala Stringa)

  • Florarobotica
  • Micro and macro behaviors. He’s talking about ants, but I’m interested in UIs. What interfaces/communication channels cause an intelligent agent to perform constructive/destructive/etc behaviors
  • Swarm Robotics: A Formal Approach
  • How do you make decisions on the macroscopic scale?
  • Fast and sloppy vs. slow and accurate. What about ‘hardware acceleration?
  • Kilobot
  • Voter models
  • Majority rule is faster diffusion than voter model, but less accurate.
  • Mixing models does affect the results. Is this explore/exploit as well?
  • Swarm performance over density is a left-skewed normal distribution? Is this because they occupy physical space?
  • Hmmm. Density is a function of dimension
  • Density adaptation. Is it true that people like density to a point? And that online always “feels” different from actual density? What does popularity ranking vs other ranking do to human behavior?
  • subCULTron
  • generic, scalable and decentralized fault detection for robot swarms
    • Robot swarms are large-scale multirobot systems with decentralized control which means that each robot acts based only on local perception and on local coordination with neighboring robots. The decentralized approach to control confers number of potential benefits. In particular, inherent scalability and robustness are often highlighted as key distinguishing features of robot swarms compared with systems that rely on traditional approaches to multirobot coordination. It has, however, been shown that swarm robotics systems are not always fault tolerant. To realize the robustness potential of robot swarms, it is thus essential to give systems the capacity to actively detect and accommodate faults. In this paper, we present a generic fault-detection system for robot swarms. We show how robots with limited and imperfect sensing capabilities are able to observe and classify the behavior of one another. In order to achieve this, the underlying classifier is an immune system-inspired algorithm that learns to distinguish between normal behavior and abnormal behavior online. Through a series of experiments, we systematically assess the performance of our approach in a detailed simulation environment. In particular, we analyze our system’s capacity to correctly detect robots with faults, false positive rates, performance in a foraging task in which each robot exhibits a composite behavior, and performance under perturbations of the task environment. Results show that our generic fault-detection system is robust, that it is able to detect faults in a timely manner, and that it achieves a low false positive rate. The developed fault-detection system has the potential to enable long-term autonomy for robust multirobot systems, thus increasing the usefulness of robots for a diverse repertoire of upcoming applications in the area of distributed intelligent automation.


Satisfy: Towards a self-learning analyzer for time series forecasting in self-improving systems

  • Hyperparameter tuning!
  • Add a meta-learning component (Forecasting module)DSCN0629
  • AutoWEKA (CASHO) – automatically defined search space <— this works!
  • Clustering on the data to try different algorithms on clusters
  • Need to spend some time looking at random forest. It seems to be high value for lower cost
  • Sante-fe institute time-series contest (set A) (set B)

Adaptive Coordination to Complete Mission Goals
Charles Walter, Sarra Alqahtani, and Rose Gamble

  • What Reasonable Guarantees Can We Make for a SISSY System? Kirstie Bellman
    • Trustworthy and knowable
    • Guarantees

Hierarchical Self-Awareness and Authority for Scalable Self-Integrating Systems
Ada Diaconescu, Barry Porter, Roberto Rodrigues Filho and Evangelos Pournaras

  • Hierarchy: Not authority and control, Not distribution. Do mean a multilevel system of different levels of abstraction with feedback loops between them
  • Levels of abstraction allow availability through loss of irrelevant information
  • Executes at different time scales
  • Abstraction goes up, feedback goes down
  • DSCN0632
  • DSCN0633

Security Issues in Self-improving System Integration – Challenges and Solution Strategies Henner Heck, Bernhard Sick and Sven Tomforde

  • Hardly manageable system structures
  • Capabilities
    • Mutual influence detection
    • Mutual dependency detection
    • Emergence detection
    • Self-reflection
  • Am I approaching the optimum?
  • Am I degraded?
  • Additional attack vectors targeting
    • self reflection
    • trust
    • mutual influence
    • adaptation

Improving Security and Interoperability of Interwoven Systems through Rigorous Selective Encapsulation of Critical Physical Resources
Phyllis Nelson

Outriggers and Training Wheels for Cooperating Systems  – Christopher Landauer

  • Kreitman’s Theorem?
  • Training wheels are constraints on a system that prevent catastrophic failures.
  • DSCN0633
  • Efficiency and robustness are direct competitors

A Concept for Proactive Knowledge Construction in Self-Learning Autonomous Systems Anthony Stein, Sven Tomforde, Ada Diaconescu, Jörg Hähner and Christian Müller-Schloer <- nice paper?

  • Reactive knowledge creation = trial and error
  • Proactive behavior = new knowledge created before it is needed
  • MLOC architectures (Multi-Layer Observer/Controller)
  • Drifting distributions leads to changes in the fitness function
  • Suggested NNMF or actually tensor factorization as a way of filling in the tensor

Aspects of Measuring and Evaluating the Integration Status of a (Sub-)System at Runtime
Christian Gruhl, Sven Tomforde, and Bernhard Sick

  • Entity and system-based evaluation

Coopetitive Soft Gating Ensemble
Jens Schreiber, Maarten Bieshaar, André Gensler, Bernhard Sick and Stephan Deist

  • Ensemble estimators?
  • Combinations of weighting and gating to do optimized selection of the best estimator from a collection of estimators that work in particular situations.
  • DSCN0635

Levels of Networked Self-awareness
Lukas Esterle (chair for next years’ SASO)  and John N.A. Brown


Phil 9.6.18

SASO 2018

IEEE (f*) proceedings list (conference should show up here)

Keynote Gábor Vásárhelyi – CollMot Robotics, Budapest, Hungary (chair A. Montresor, room: FBK – Sala Stringa)

  • Multi-level hierarchy in pigeons (gps), wild horses (drone video)
  • Tight turns with pigeons rearrange order. Gentle turns maintain structure
  • Universal rules of collective motion
  • evolution is the best optimizer that they have found
  • Hierarchy aids emergent optimization
  • Drones are exploding in youth. Why? Fashion? Like AI in the ’70’s. Will it then crash?
  • Salable in size and speed
  • Tolerate noise, delay, and error tolerant
  • Controllable meta-unit
  • Drones need to be
    • Extensable
    • real-time
    • sensors
    • communication
    • simulation for dev and debugging
    • algorithms
  • Enemies
    • delay
    • noise
    • inertia,
    • finite acceleration
    • communication errors
    • environment?
  • Communication delays bring self-exited oscillation (repulsion-attraction temporal imbalance)
  • Evolution is the second-best optimization function for all problems
  • Solving decentralized traffic control. A sort of inverse flocking issue
  • Anisotropic repulsion
  • Selective alignment
  • Publications
  • Smaller and lighter (more maneuverable) flocks are more egalitarian. Heavier, smaller groups have hierarchies because there can be trust.

Rule-based utility-driven???
Sona Ghahremani, Christian Medeiros Adriano and Holger Giese

  • Another hyperparameter approach. This one seems quite good
  • Handles
    • Non-linearities
    • Uncertainty from dynamic archetecture
    • Black box models
  • How to select the right model for a class of utility functions prior to deployment
  • Graph grammar rules
  • mRUBiS
    • mRUBiS as an exemplar for model-based architectural self-adaptation is now available at GitHub. It provides a simulator and architectural runtime model of mRUBiS to develop, evaluate, and compare self-adaptation solutions.

Model-Driven Elasticity Control for Multi-Server Queues Under Traffic Surges in Cloud Environments
Venkat Tadakamalla and Daniel Menasce

A Temporal Model for Interactive Diagnosis of Adaptive Systems
Ludovic Mouline, Amine Benelallam, Francois Fouquet, Johann Bourcier and Olivier Barais

  • Might be a way of evaluating fitness tests?
  • At least keep track of the last N (or whole history?) of chromosome so that the principal components can be analyzed.
  • By comparing the states over time, we could find the states adjacent to problem states?

Towards Generic Adaptive Monitoring
Thomas Brand and Holger Giese

  • Goal adaptation (fitness test adaptation?)

Implementing Feedback for ABM Meta-Modeling
Karan Budhraja and Tim Oates

PINCH: Self-Organized Context Neighborhoods for Smart Environments
Chenguang Liu, Christine Julien and Amy L. Murphy

  • Neighborhood mesh networks
  • everything is stored in a small beacon that is cooperatively shared in the neighborhood

A QoS-aware Adaptive Mobility Handling Approach for LoRa-based IoT Systems
Robbe Berrevoets and Danny Weyns

Reins-MAC: Firefly Inspired Communication Scheduling for Reliable Low-Power Wireless
Matteo Ceriotti and Amy L. Murphy

Phil 9.5.18

Marco Barbina – Leonardo SpA – Italy (chair E. Di Nitto, room: FBK – Sala Stringa)

  • Sensors on UAVs
    • Electro-optic
    • Radar-based
    • Hyper-spectral(?)
  • Sensors are not the problem, bandwidth is
  • Selecting the relevant information
  • Attention – which parts of the signal are important
    • Autoencoder NN – it’s like dropout?
    • Split domain stacked cnn. The front half (encoder) NN encodes the data and transmits the “compressed” data from the hidden layer to the back half of the NN.

Goal-aware Team Affiliation in Collectives of Autonomous Robots
Lukas Esterle

  • Teamwork in nature
  • Online multi-task k-assignment
  • stampedes!
  • Observer/follower == explore/exploit
  • They did use the multi-armed bandit framework as a learning system.
  • Real world implementation is intruder detection and response

Optimizing Transitions Between Abstract ABM Demonstrations (video presentation)
Brian Seipp, Karan Budhraja and Tim Oates

  • What is the optimal path that connects two behaviors
  • Traverse a behavior coordinate space
  • Getting from a start image to a goal image. The image can “mean” many things. Agents provide the mapping?
  • Intermediate states are shown
  • Noise added to guarantee convergence?

Self-organized Resource Allocation for Reconfigurable Robot Ensembles
Julian Hanke, Oliver Kosak, Alexander Schiendorfer and Wolfgang Reif

  • Search – determine dangers
  • Continual observe
  • React
  • This is a fitness space analysis based on optimal configuration of robots for a given task. It’s kind of a distributed traveling salesman problem combined with a market and a bidder as a way of merging the distributed solution. I wonder if it is computationally better than a GA solution. It could be a way of looking at hyperparameter tuning?
  • Not clear that it can handle a floating point value, like fuel level
  • The goal is for the allocation to be fast

Panel: The Future of Autonomic Computing and Self-* Systems:

  • Kristie Bellman – some success but in isolated, hard to define areas. Build foundation of principals that can be used to frame development
  • Ada Diaconescu – Efficiency-oriented or creativity-oriented self* systems
  • Simon Dobson – We have turned a lot of fields into computation. But – AI told us nothing about how people play Go. We’re trying to build, not describe
  • Xiaohui (Helen) Gu – AIOps, bridging the gap between academia and industry
  • Arif Merchant – (Google) Missing holistic, end-to-end solutions. First and follow-on principals.
  • Danny Weyns – Be less ad-hoc and more scientific. Address threats to validity. Solid foundations.
  • I talked a bit about the continuum from the low-hanging fruit of autonomic systems to the difficult task of creating self-aware systems.

A Macro-Level Order Metric for Self-Organizing Adaptive Systems
David King and Gilbert Peterson

  • critical and stable states, based on local and global entropy
  • I think that he is talking about “personalized” entropy rather than “local” entropy.

A Self-Organized Learning Model for Anomalies Detection: Application to Elderly People
Nicolas Verstaevel, Carole Bernon, Jean-Pierre Georgé and Marie-Pierre Gleizes

  • Real time detection of abnormal behaviors by using feedback from the medical staff.
  • Detect anomalies through a linear regression of disparity values
  • Is this another hyperparameter system? It could be looking for sensitivity of the system to hyperparameter

Risk-based Testing of Self-Adaptive Systems using Run-Time Predictions
André Reichstaller and Alexander Knapp

  • Uses machine learning to explore the space that includes errors. It is a way of finding major(?) features in the state space

Play with Generated Adversarial Networks (GANs) in your browser!

  • First, we’re not visualizing anything as complex as generating realistic images. Instead, we’re showing a GAN that learns a distribution of points in just two dimensions. There’s no real application of something this simple, but it’s much easier to show the system’s mechanics. For one thing, probability distributions in plain old 2D (x,y) space are much easier to visualize than distributions in the space of high-resolution images.

Phil 9.4.18

SASO 2018

This works: eduroam wireless network Setup for Windows 10: wi-fi-eduroam-windows-10-en

Keynote 1 Paula ??? Privacy

  • Privacy is a right of a human being that must be protected
  • Privacy as soft paternalism Nudges for Privacy and Security ACM 2017
  • Mangrove societies
  • USACM EUACM ethics statements
  • “big data protection ecosystems that shall involve developers, businesses, regulators and individuals in order to provide ‘ future-oriented regulation’, accountable controllers’, ‘privacy conscious engineering’, and ’empowered individuals'”.
  • Lifting humans to be actors in the digital world by becoming autonomous systems that interact “au pair” with the rest of the digital world
  • Floridi – Soft ethics in the governance of the digital. Philosophy and Technology March 2018
  • Soft ethics is associated to individuals and hard ethics to systems, i.e. autonomous cars
  • The two need to combine when an individual and a system interact

David Kurka Disobedience as a mechanism of change

  • The Iron Law of Oligarchy
  • The social concept of disobedience
  • Phase plots ???
  • All the work is currently based on global understandings of fairness? Future work is will look at subjective values within populations

Shuyue Hu –

  • Coordination games induce social and individual interest
  • Iterated prisoner’s dilemma but no strategy. Agents can decide to C or D based on the last round

Jeremy Pitt & Josiah Ober (Stanford)- Democracy by design

  • Elinor Ostrom
  • SimDemopolis (Specification and implementation)
  • Brute facts and institutional facts
  • Immutable and mutable rules
  • Gini index keeps popping up. Use this math to reformat stampede/flock/nomad chart?

A Multi-Agent Elasticity Management Based On Multi-Tenant Debt Exchanges
Carlos Mera-Gómez, Francisco Ramírez, Rami Bahsoon and Rajkumar Buyya

  • Stable matching approach (2012 Nobel economics)
  • Burlap reinforcement learning
    • The Brown-UMBC Reinforcement Learning and Planning (BURLAP) java code library is for the use and development of single or multi-agent planning and learning algorithms and domains to accompany them. BURLAP uses a highly flexible system for defining states and and actions of nearly any kind of form, supporting discrete continuous, and relational domains. Planning and learning algorithms range from classic forward search planning to value function-based stochastic planning and learning algorithms. Also included is a set of analysis tools such as a common framework for the visualization of domains and agent performance in various domains.

Forecasting Models for Self-Adaptive Cloud Applications: A Comparative Study
Vladimir Podolskiy, Anshul Jindal, Michael Gerndt and Yury Oleynik

  • Time series prediction algorithms
    • ARIMA –
    • GARCH – variance in the error term
    • Support Vector Regression
    • Singular Spectrum Analysys
  • Interval Accuracy score



Phil 8.31.18

7:00 – 5:00 ASRC MKT

  • The lightning round slides are in!
  • Get Speaker – done
  • Get posters – done
  • Haircut – done
  • drop off DME/KLR – done
  • Under Pressure response – done, I think?
  • upload ML excel files (done) to play around with graph laplacians some more – done
  • Print out two travel packets – done
  • create shared itinerary document – started. Aaron needs to finish his part
  • From KQED Silicon Valley Conversations The Future of Music: Computer or Composer
    • Ge Wang is an Associate Professor at Stanford University in the Center for Computer Research in Music and Acoustics (CCRMA). He specializes in the art of computer music design — researching programming languages and interactive software design for music, interaction design, expressive mobile music, new performance ensembles (laptop orchestra and mobile phone orchestra), human-computer interaction, visualization (sndpeek), music game design, aesthetics of technology-mediated design, and methodologies for education at the intersection of art, engineering, and design.
    • Doug Eck is a research scientist working on Magenta, a research project exploring the role of machine learning in the process of creating art and music. Primarily this involves developing new deep learning and reinforcement learning algorithms for generating songs, images, drawings, and other materials. But it’s also an exploration in building smart tools and interfaces that allow artists and musicians to extend (not replace!) their processes using these models. Started by me in 2016, Magenta now involves several researchers and engineers from the Google Brain team as well as many others collaborating via open source. Aside from Magenta, I’m working on sequence learning models for summarization and text generation as well new ways to improve AI-generated content based on user feedback.
    • Amy X Newburg has been developing her own brand of irreverently genre-crossing works for voice, live electronics and chamber ensembles for over 25 years, known for her innovative use of live looping technology with electronic percussion, her 4-octave vocal range and her colorful — often humorous — lyrics. One of the earliest performers to work with live digital looping, Amy has presented her solo “avant-cabaret” songs at such diverse venues as the Other Minds and Bang on a Can new music festivals, the Berlin International Poetry Festival, the Wellington and Christchurch Jazz Festivals (New Zealand), the Warsaw Philharmonic Hall, electronic music festivals, colleges, rock clubs and concert halls throughout the U.S. and abroad.
  • Teens, Social Media & Technology 2018
    • YouTube, Instagram and Snapchat are the most popular online platforms among teens. Fully 95% of teens have access to a smartphone, and 45% say they are online ‘almost constantly’
  • Aaron found this: Density-functional fluctuation theory of crowds
    • A primary goal of collective population behavior studies is to determine the rules governing crowd distributions in order to predict future behaviors in new environments. Current top-down modeling approaches describe, instead of predict, specific emergent behaviors, whereas bottom-up approaches must postulate, instead of directly determine, rules for individual behaviors. Here, we employ classical density functional theory (DFT) to quantify, directly from observations of local crowd density, the rules that predict mass behaviors under new circumstances. To demonstrate our theory-based, data-driven approach, we use a model crowd consisting of walking fruit flies and extract two functions that separately describe spatial and social preferences. The resulting theory accurately predicts experimental fly distributions in new environments and provides quantification of the crowd “mood”. Should this approach generalize beyond milling crowds, it may find powerful applications in fields ranging from spatial ecology and active matter to demography and economics.
    • Here’s an interesting part: The DFFT analysis that we present is particularly powerful because it separates the influence of the environment on agents from interactions among those agents. 
      • This implies that it should (could? might?) be possible to calculate a social/environmental ratio for individual agents. High environmental are nomadic. High social are stampede-prone. Need to dig in further.
  • Mechanical Vibrations and Waves » Lecture 4: Coupled Oscillators, Normal Modes
    Lecture 4: Coupled Oscillators, Normal Modes (MIT opencourseware)

    • Prof. Lee analyzes a highly symmetric system which contains multiple objects. By physics intuition, one could identify a special kind of motion – the normal modes. He shows that there is a general strategy for solving the normal modes.
      • Every part of the system is oscillating at the same frequency and the same phase
      • Stopped at 42:07 to take a break. I think this is the right track though. Download this for the plane?
  • Chapter on normal modes

Phil 8.30.18

7:00 – 5:00  ASRC MKT

  • Target Blue Sky paper for iSchool/iConference 2019: The chairs are particularly looking for “Blue Sky Ideas” that are open-ended, possibly even “outrageous” or “wacky,” and present new problems, new application domains, or new methodologies that are likely to stimulate significant new research. 
  • I’m thinking that a paper that works through the ramifications of this diagram as it relates to people and machines. With humans that are slow responding with spongy, switched networks the flocking area is large. With a monolithic densely connected system it’s going to be a straight line from nomadic to stampede. Nomad-Flocking-Stampede2
    • Length: Up to 4 pages (excluding references)
    • Submission deadline: October 1, 2018
    • Notification date: mid-November, 2018
    • Final versions due: December 14, 2018
    • First versions will be submitted using .pdf. Final versions must be submitted in .doc, .docx or La Tex.
  • More good stuff on BBC Business Daily Trolling for Cash
    • Anger and animosity is prevalent online, with some people even seeking it out. It’s present on social media of course as well as many online forums. But now outrage has spread to mainstream media outlets and even the advertising industry. So why is it so lucrative? Bonny Brooks, a writer and researcher at Newcastle University explains who is making money from outrage. Neuroscientist Dr Dean Burnett describes what happens to our brains when we see a comment designed to provoke us. And Curtis Silver, a tech writer for KnowTechie and ForbesTech, gives his thoughts on what we need to do to defend ourselves from this onslaught of outrage.
  • Exposure to Opposing Views can Increase Political Polarization: Evidence from a Large-Scale Field Experiment on Social Media
    • Christopher Bail (Scholar)
    • There is mounting concern that social media sites contribute to political polarization by creating “echo chambers” that insulate people from opposing views about current events. We surveyed a large sample of Democrats and Republicans who visit Twitter at least three times each week about a range of social policy issues. One week later, we randomly assigned respondents to a treatment condition in which they were offered financial incentives to follow a Twitter bot for one month that exposed them to messages produced by elected officials, organizations, and other opinion leaders with opposing political ideologies. Respondents were re-surveyed at the end of the month to measure the effect of this treatment, and at regular intervals throughout the study period to monitor treatment compliance. We find that Republicans who followed a liberal Twitter bot became substantially more conservative post-treatment, and Democrats who followed a conservative Twitter bot became slightly more liberal post-treatment. These findings have important implications for the interdisciplinary literature on political polarization as well as the emerging field of computational social science.
  • Setup gcloud tools on laptop – done
  • Setup Tensorflow on laptop. Gave up un using CUDA 9.1, but got tf doing ‘hello, tensorflow’
  • Marcom meeting – 2:00
  • Get the concept of behaviors being a more scalable, dependable way of vetting information.
    • Eg Watching the DISI of outrage as manifested in trolling
      • “Uh. . . . not to be nitpicky,,,,,but…the past tense of drag is dragged, not drug.”: An overview of trolling strategies
        • Dr Claire Hardaker (Scholar) (Blog)
          • I primarily research aggression, deception, and manipulation in computer-mediated communication (CMC), including phenomena such as flaming, trolling, cyberbullying, and online grooming. I tend to take a forensic linguistic approach, based on a corpus linguistic methodology, but due to the multidisciplinary nature of my research, I also inevitably branch out into areas such as psychology, law, and computer science.
        • This paper investigates the phenomenon known as trolling — the behaviour of being deliberately antagonistic or offensive via computer-mediated communication (CMC), typically for amusement’s sake. Having previously started to answer the question, what is trolling? (Hardaker 2010), this paper seeks to answer the next question, how is trolling carried out? To do this, I use software to extract 3,727 examples of user discussions and accusations of trolling from an eighty-six million word Usenet corpus. Initial findings suggest that trolling is perceived to broadly fall across a cline with covert strategies and overt strategies at each pole. I create a working taxonomy of perceived strategies that occur at different points along this cline, and conclude by refining my trolling definition.
        • Citing papers
  • FireAnt (Filter, Identify, Report, and Export Analysis Toolkit) is a freeware social media and data analysis toolkit with built-in visualization tools including time-series, geo-position (map), and network (graph) plotting.
  • Fix marquee – done
  • Export to ppt – done!
    • include videos – done
    • Center title in ppt:
      • model considerations – done
      • diversity injection – done
  • Got the laptop running Python and Tensorflow. Had a stupid problem where I accidentally made a virtual environment and keras wouldn’t work. Removed, re-connected and restarted IntelliJ and everything is working!

Phil 8.29.18

7:00 – 4:30 ASRC MKT

  • This Is How Russian Propaganda Actually Works In The 21st Century (plus Kate Starbird’s twitter thoughts)
    • The Russian government discreetly funded a group of seemingly independent news websites in Eastern Europe to pump out stories dictated to them by the Kremlin, BuzzFeed News and its reporting partners can reveal.
  • How Right Wing is Right Wing Populism? Using multilingual CNNs on party manifestos.
    • Right wing populist parties in Europe are clearly different from other right wing parties in their rhetoric and electoral appeal. Some observers see substantive differences between right wing populists and other right wing parties, with populists supporting the welfare state and gender equality more than other right wing parties, often as part of an anti-immigration and anti-Muslim agenda. We test this claim using novel data produced by a multilingual convolutional neural net on political party platforms for the years 1990 to 2015 from the Manifesto Corpus. We find no systematic differences between right wing populists and non-populists on support for welfare and gender equality, though there is some evidence that more successful populists are more centrist.
  • Need to write up a 4 page blue sky paper for the 2019 iConference in DC
  • Realized that the poster had two herding DTW charts on the poster. Fixed and sent back. Hopefully it will get reprinted in time…
  • Uploaded the edited version and added them to the online presentation. Also saved out the mp4 files to use in the ppt version
  • Back to working on speech recognition. I’ve done a bunch of things that I’m documenting before I see if anything helped.
  • TL;DR – after much flailing, I found a page that actually helped. It’s a how-to (rather than quickstart) guide that includes a variety of interfaces including gcloud, Java and Python. And the gcloud command worked like a charm! All the flailing below is just for documentation on what NOT to do. Here’s what worked:
    PS D:\Development\Sandboxes\MapsFromPodcasts> gcloud ml speech recognize D:\Development\Sandboxes\MapsFromPodcasts\brook
    lyn.flac --language-code='en-US'
      "results": [
          "alternatives": [
              "confidence": 0.98360395,
              "transcript": "how old is the Brooklyn Bridge"

    Note that the audio file is the same as the one in the examples and is available from Google here:

  • For historical documentation of my flailing
    • First I opened a new Powershell window and re-ran the commands. Yup: Capture
    • Then I stumbled on the SDK support page and found this link to what may be the answer to the question on stackoverflow. CaptureIt says to run
      gcloud auth application-default login --scopes=,
    • Which I did, which caused a lot of things to happen Capture
    • First, I’m really wondering about this: To generate an access token for other uses, run: gcloud auth application-default print-access-token. This is used in both commands, si I’m wondering what it’s actually doing. What’s happening to this generated  token? is it being stored on my machine?
    • Second, it looks like I need to point at the [C:\Users\philip.feldman\AppData\Roaming\gcloud\application_default_credentials.json] file rather than the one in the project. That or copy to the dev location. I’m trying the former Capture
    • Then, I got this again ( Capture
    • Lastly, I upgraded because it said I could. Nothing works yet, so why not? Capture
    • That brought up a window with all this info:
      Your current Cloud SDK version is: 213.0.0
      You will be upgraded to version: 214.0.0
      │        These components will be updated.        │
      │           Name           │  Version   │   Size  │
      │ Cloud SDK Core Libraries │ 2018.08.24 │ 8.3 MiB │
      │ gcloud cli dependencies  │ 2018.08.24 │ 2.4 MiB │
      │                 These components will be installed.                 │
      │            Name            │       Version       │       Size       │
      │ Bundled Python             │                     │                  │
      The following release notes are new in this upgrade.
      Please read carefully for information about new features, breaking changes,
      and bugs fixed.  The latest full release notes can be viewed at:
      214.0.0 (2018-08-28)
        Breaking Changes
            ■ **(Cloud Bigtable)** Modified the arguments accepted by cbt
              createappprofile and cbt updateappprofile in the following ways:
              ≡ Removed etag argument from createappprofile.
              ≡ Renamed allow-transactional-writes option as transactional-writes.
              ≡ Added a force option to ignore warnings.
            ■ **(Cloud Bigtable)** Modified the specification for routing policies.
              A routing policy can be either "route-any" (previously of
              "multi_cluster_routing_use_any") or "route-to=".
            ■ **(Compute Engine)** Deprecated gcloud compute interconnects
              attachments create. Please use gcloud compute interconnects attachments
              dedicated create instead.
            ■ **(Compute Engine)** Removed deprecated --mode flag from gcloud
              compute networks create. Use --subnet-mode instead.
            ■ **(Compute Engine)** Removed deprecated gcloud compute networks
              switch-mode command. Use gcloud compute networks update
              --switch-to-custom-mode instead.
            ■ **(Compute Engine)** Removed deprecated gcloud compute xpn command
              group. Use gcloud compute shared-vpc instead.
        Cloud Bigtable
            ■ Restored the output of the cbt count command that was inadvertently
              removed in the previous release.
        Cloud Datalab
            ■ Updated the datalab component to the 20180820 release. Released
              changes are documented in its tracking issue at
        Cloud Dataproc
            ■ Added SCHEDULED_DELETE column to gcloud beta dataproc clusters list
              command output.
        Cloud Datastore Emulator
            ■ Released Cloud Datastore Emulator version 2.0.2.
              ≡ Improved backward compatibility with App Engine local development
                by keeping auto generated indexes in index file generated from
                previous runs.
        Cloud Functions
            ■ Promoted --runtime flag of gcloud functions deploy to GA.
        Compute Engine
            ■ Promoted the following flags to GA:
              ≡ --network-tier of gcloud compute <addresses|forwarding-rules>
              ≡ --default-network-tier of gcloud compute project-info update
              ≡ --network-tier of gcloud compute instances
              ≡ --network-tier of gcloud compute instance-templates create
            ■ Promoted gcloud compute instances simulate-maintenance-event to GA.
            ■ Promoted <get|set>-iam-policy and <add|remove>-iam-policy-bindings to
              beta in the following commands groups:
              ≡ gcloud compute sole-tenancy node-groups
              ≡ gcloud compute sole-tenancy node-templates
        Kubernetes Engine
            ■ Promoted --disk-type flag of gcloud container <clusters|node-pools>
              create to GA.
            ■ Promoted --default-max-pods-per-node flag of gcloud container
              clusters create to beta.
            ■ Promoted --max-pods-per-node flag of gcloud container node-pools
              create to beta.
            ■ Modified --monitoring-service flag of gcloud containers clusters
              update to enable Google Cloud Monitoring service with Kubernetes-native
              resource model.
            ■ Modified --logging-service flag of gcloud containers clusters update
              to enable Google Cloud Logging service with Kubernetes-native resource
            ■ Modified output of gcloud beta container clusters list for DEGRADED
              clusters to include reason for degradation.
            ■ Added --enable-private-nodes and --enable-private-endpoint to gcloud
              beta container clusters create.
            ■ Deprecated --private-cluster flag of gcloud beta container clusters
              create; use --enable-private-nodes instead.
          Subscribe to these release notes at
      Do you want to continue (Y/n)?  Y
      ╠═ Creating update staging area                             ═╣
      ╠═ Uninstalling: Cloud SDK Core Libraries                   ═╣
      ╠═ Uninstalling: gcloud cli dependencies                    ═╣
      ╠═ Installing: Bundled Python                               ═╣
      ╠═ Installing: Cloud SDK Core Libraries                     ═╣
      ╠═ Installing: gcloud cli dependencies                      ═╣
      ╠═ Creating backup and activating new installation          ═╣
      Performing post processing steps...done.
      Update done!
      To revert your SDK to the previously installed version, you may run:
        $ gcloud components update --version 213.0.0
      Press any key to continue . . .
    • So now lets see what happens with a restarted PowerShell
    • Nope, same problem. I also tried deleting the environment variable completely and the behavior is the same. So I don’t think that the file with the data is being sent? Capture
    • Interesting, the app-roaming file is not the same as the file that google had me generate for the text recognition getting started page: Capture

Phil 8.29.18

7:00 – 4:30 ASRC MKT

  • Editing videos
  • Need to think about short CHI paper about designing for culture/robot interactions. The trolly problem at scale? How would the sim be set up? The amount of randomness at the initial condition? Stiffness vs. connectivity? Beleif space is still important and is actually used as a concept in path planning
  • Visual Exploration and Comparison of Word Embeddings
    • Word embeddings are distributed representations for natural language words, and have been wildly used in many natural language processing tasks. The word embedding space contains local clusters with semantically similar words and meaningful directions, such as the analogy. However, there are different training algorithms and text corpora, which both have a different impact on the generated word embeddings. In this paper, we propose a visual analytics system to visually explore and compare word embeddings trained by different algorithms and corpora. The word embedding spaces are compared from three aspects, i.e., local clusters, semantic directions and diachronic changes, to understand the similarity and differences between word embeddings.
  • Much work on slides
  • Can’t get Google to recognise my account?
    curl.exe -H "Content-Type: application/json" -H "Authorization: Bearer "$(gcloud auth application-default print-access-token) -d @sync-request.json
    curl: (6) Could not resolve host: ya29.c.EloHBu32-0nBAqimi1Zumlot6rjGtGpUk27qTTESRLW4vtd1LY4ihxBIesU3ga-kmwCaM7YZS-JRo_KNjaC_bj13dWazBcKr4YtAEQYFzSpSBx3DwdS46DTt0bg
      "error": {
        "code": 403,
        "message": "The request is missing a valid API key.",
        "status": "PERMISSION_DENIED"

    No idea what host: ya29.c.EloHBu32-0nBAqimi1Zumlot6rjGtGpUk27qTTESRLW4vtd1LY4ihxBIesU3ga-kmwCaM7YZS-JRo_KNjaC_bj13dWazBcKr4YtAEQYFzSpSBx3DwdS46DTt0bg is

  • Found a problem with the poster. There are two herding DTW charts. Must be reprinted

Phil 8.24.18

7:00 – 4:00 ASRC MKT

  • Make more obvious the Inadvertent Social Information and Digital ISI
    • ISI
      • Trails
      • Visual clustering
      • Behavior around the commons (waterholes)
      • Presence of young
      • Mating behavior
      • etc.
    • DISI
      • Words and their overall source (Social media, website content, contributor content, auto-generated, etc)
      • Votes (likes, kudos, karma points)
      • Money (site income, blockchain ledger)
      • Linking (href, retweet, share)
      • Images & videos
  • Work more on behavior patterns of humans and animals
    • Highly organized (soccer match singing, marching, mass dancing events)
    • Wildebeest feeding, defending,migrating and stampeding
  • AutoKeras is a GitHub project that uses the ENAS algorithm. It can be installed using pip. Since it’s written in Keras it’s quite easy to control and play with, so you can even dive into the ENAS algorithm and try making some modifications. If you prefer TensorFlow or Pytorch, there’s also public code projects for those here and here!
  • From Zeynep’s twitter
    • So, Russian trolls amplified divisive content and helped spread vaccine misinformation.  Look, the challenge before us is to redefine *critical thinking* to include figuring out what to believe, not just how to be skeptical. Personal and institutional.
    • Weaponized Health Communication: Twitter Bots and Russian Trolls Amplify the Vaccine Debate
      •  Whereas bots that spread malware and unsolicited content disseminated antivaccine messages, Russian trolls promoted discord. Accounts masquerading as legitimate users create false equivalency, eroding public consensus on vaccination.
  • Trying to decode podcasts. Here’s my test , and here are the results from Google speech-to-text:
    • We were talking about the choices of who’s you can keep two of these three, I guess Adonis Alexander is along for the ride, huh? I thought I was about to I didn’t know I haven’t I haven’t sent it to him. Well, has he been out there? They might missing some guys got to hand. I kept thinking like if to say, they weren’t having these injuries. Like if they have like us to say, okay, but they have these reason iron Marshall and maybe he maybe he’s not available week one, but they don’t want to put them on IR prn’s things up. So maybe they have to add another running back like you so you have to create a roster spot I could imagine this is just speculation Alexander. Somehow gets the mysterious injury to put them on I are clearly my keys ready, right and they they would have five cornerbacks otherwise and you know, yeah, if you’re not going to be ready to go, but you may have to you know, go get okay. Yeah. I mean the he’s he’s a guy that I think is on based on these the way the wrong.
    • It’s pretty good as long as people aren’t stepping over each other verbally.
    • Good enough to try, I guess. Noisy data is life, right? Look for the bigger signal.
  • Here’s my current plan. It’s a half-assed first approach, but it should provide some insight.
    1. Download a season of a sports podcast and put each podcast into it’s own document Here’s the tutorial for Speech-to-text with REST
    2. Use Corpus Manager to convert, using BOW and create an ignore list for common words like “the”
    3. then read all the docs into LMN
    4. Then set the weight of each successive document (in time) so that its top
    5. Take the top ten words and save them to a file
    6. Try building a map

Phil 8.23.18

7:00 – 5:30 ASRC MKT


  • Slides
    • Groups/tribes stay the same, but the topics change
    • Past polarizing topics:
      • Confederate statues
      • Kneeling for the national anthem
      • #blacklivesmatter
      • Hoodies
      • Crack cocaine
      • 1968 Olympics Black Power salute
      • Alabama bus boycott
    • Stiffening a group creates a stampede (In-group high SIH)
    • Adding group-invisible diversity disrupts the velocity and direction of a stampede
    • Arendt/Moscovici slide “So we’re doomed, right! Except…”
    • See what velocity of the disrupted stampede looks like
  • Why Trump Supporters Believe He Is Not Corrupt
    • The answer may lie in how Trump and his supporters define corruption. In a forthcoming book titled How Fascism Works, the Yale philosophy professor Jason Stanley makes an intriguing claim. “Corruption, to the fascist politician,” he suggests, “is really about the corruption of purity rather than of the law. Officially, the fascist politician’s denunciations of corruption sound like a denunciation of political corruption. But such talk is intended to evoke corruption in the sense of the usurpation of the traditional order.”
  • Climate science proposals are being reviewed by Ryan Zinke’s old football buddy. Seriously.
    • But what if the corruption isn’t hidden at all, but right out in the open? What if, when it’s identified, the perpetrator doesn’t apologize, or demonstrate any remorse or shame, and there’s no punishment? What then? We don’t really have good narratives around what happens in that situation, which is why the Trump administration so often leaves us sputtering and gawking. It can’t just be a motley collection of incompetent grifters, each misruling their own little fiefdom, trying to stay in their boss’s good graces, succeeding less through wits than a congenital lack of shame and the unstinting institutional support of GOP donors. Can it?