Showing posts with label MTD. Show all posts
Showing posts with label MTD. Show all posts

Thursday, January 11, 2024

Activate the Self Replicate


This day is a seminal moment in the life of this weblog -- this is the closest we've ever come to seeing the word "self-replicate" in the news.

They refer to it as "digital replication", "living individual replicas", and "digital doppelgänger" but the concept has become a tightly arranged crystal of thought-parts. The idea of multiple, simultaneous instances of a single original self is now well-formed. 

What's better is that this was published in The Conversation, which is a general audience magazine not a science journal, or a science fiction story for that matter:

Image credit: Try not to do this often but this is not the first time this image has been used as a thumbnail: René Magritte, Not to Be Reproduced (La reproduction interdite, 1937


AI clones made from user data pose uncanny risks
Jun 2023, phys.org

This mirror image of an individual created by artificial intelligence is referred to as an "AI clone." Our study dives into the murky waters of what these AI clones could mean for our self-perception, relationships and society. 

We presented 20 participants with eight speculative scenarios involving AI clones. The participants were diverse in ages and backgrounds, and reflected on their emotions and the potential impacts on their self-perception and relationships.

What we fear:
  • Digital counterparts could exploit and displace their identity
  • Threat of identity fragmentation
  • Lastly, participants expressed concerns about what we described as "living memories." This relates to the danger posed when a person interacts with a clone of someone they have an existing relationship with. Participants worried that it could lead to a misrepresentation of the individual, or that they would develop an over-attachment to the clone, altering the dynamics of interpersonal relationships.

via University of British Columbia: Patrick Yung Kang Lee et al, Speculating on Risks of AI Clones to Selfhood and Relationships: Doppelganger-phobia, Identity Fragmentation, and Living Memories, Proceedings of the ACM on Human-Computer Interaction (2023). DOI: 10.1145/3579524

Note: As the volume of personal data we generate continues to grow, so too does the fidelity of these AI clones in replicating our behavior.

Word watch: "user-generated data expiration strategies"


Multiple 'selves' of modular agents boost AI learning
Jul 2023, phys.org

The book writes itself: 

A study comparing reinforcement learning approaches used in single AI agent and modular multi-AI agent systems -- trained deep reinforcement learning agents in a simple survival game were trained to seek various resources hidden around the field and to maintain sufficient supply levels to prevail.

One agent, seen as the "unified brain" or "self," operated in standard fashion, taking a step-by-step approach to evaluate each objective and, through trial-and-error, learning what the best solutions are each step of the way.

The modular agent, however, relied on input from sub-agents that had more narrowly defined goals and had their own unique experiences, successes and failures. Once input from the multiple modules were assessed in a single "brain," the agent made choices on how to proceed.

The singular agent achieved the game's goals after 30,000 training steps. The modular agent learned faster, making significant progress after only 5,000 learning steps.

via Princeton Neuroscience Institute: Zack Dulberg et al, Having multiple selves helps learning agents explore and adapt in complex changing worlds, Proceedings of the National Academy of Sciences (2023). DOI: 10.1073/pnas.2221180120

Post Script:
Playing Tetris by committee (the Octopad, an eight player game controller)
May 2019, BBC
May 2019, phys.org
July 2019, University of Washington News

Bonus: the monolithic approach struggled with "the curse of dimensionality" -- the exponentially spiraling growth of options as the complexity of the environment was increased.


Tuesday, January 9, 2024

Anthropo Watcher


The Anthropocene: Canadian lake mud 'symbolic of human changes to Earth'
Jul 2023, BBC News

Anthropo

Crawford Lake, a small body of water in Ontario, Canada, is being put forward as the location that best records humanity's impacts on Earth.

Scientists are trying to define a new geological time period to recognise the changes we've made to the planet, and Crawford is their model example. Its sediments have captured fallout from intense fossil fuel burning, and even the plutonium from bomb tests. The muds would be symbolic of the onset of a proposed Anthropocene Epoch.

A formal start date has also been identified - the 1950s. This decade marks the beginning of the "Great Acceleration", when the human population and its consumption patterns suddenly speeded up. It coincides with the spread of ubiquitous "techno materials", such as aluminium, concrete and plastic.

via Anthropocene Working Group (AWG)

Completely unrelated image credit: Perovskite light emitting diode quantum random number generation - Magnus Johansson, Linköping University, and Thor Labs of NJ - 2023

Friday, July 29, 2022

Completely Unrelated


Regent honeyeater: Endangered bird 'has forgotten its song' 
Mar 2021, BBC News

"They don't get the chance to hang around with other honeyeaters and learn what they're supposed to sound like," explained Dr Ross Crates.

During this painstaking search, he started to notice birds that were "singing weird songs".
He recalled: "They didn't sound anything like a regent honeyeater - they sounded like different species."

Songbirds learn their songs the same way that humans learn how to speak.

via: Crates Ross, et al. 2021. Loss of vocal culture and fitness costs in a critically endangered songbird. Proc. R. Soc. B. 2882021022520210225http://doi.org/10.1098/rspb.2021.0225

"Even small numbers of extricates could synthesize a basic cognition support network. The outer fringe was full of such partially configured quasi-humans, fumbling together towards a mutant existence."
-Mass Transference Device, 2010

Image credit: An illustration of the chain-reaction process that underlies the photon avalanching mechanism Columbia Engineering researchers have realized in their nanoparticles. In this process, the absorption of a single low-energy photon sets off a chain reaction of energy transfers and further absorption events that result in many highly excited ions within the nanoparticle, which then release their energy in the intense emission of many higher-energy photons. Credit: Mikołaj Łukaszewicz/ Polish Academy of Sciences


Researchers find 12 semidetached mass-transfer massive binaries in galaxy M31
Apr 2022, phys.org

In this study, the researchers have found that the relationship between the mass ratio and the fill-out factor of the primary star reveals that they are in the stage of slow mass transfer from less massive components to their companions with the reversed mass ratio.

via Chinese Academy of Sciences: F.-X. Li et al, Semidetached Mass-transfer Massive Binaries in the Nearby Galaxy M31, The Astronomical Journal (2022). DOI: 10.3847/1538-3881/ac5685

Sunday, May 5, 2013

Recombinant Memetics and Narrative Networks


Structural Forces
Macdonald Educational (A Unit For Teachers), published for the Schools Council by MacDonald Educational Ltd., London, 1972-1973

Recombinant Memetics
This one's quite speculative, and it's technically speaking still in the proto-science phase. But it'll only be a matter of time before scientists get a better handle on the human noosphere (the collective body of all human information) and how the proliferation of information within it impacts upon virtually all aspects of human life.

Similar to recombinant DNA (in which different genetic sequences are brought together to create something new), recombinant memetics is the study of how memes (ideas that spread from person to person) can be adjusted and merged with other memes and memeplexes (a cohesive collection of memes, like a religion) for beneficial or ‘socially therapeutic' purposes (such as combating the spread of radical and violent ideologies). This is similar to the idea of 'memetic engineering' — which philosopher Daniel Dennett suggested could be used to maintain cultural health. Or what DARPA is currently doing via their ‘narrative control' program.
11 Emerging Scientific Fields That Everyone Should Know About, George Dvorsky 27 Feb2013


The Narrative of a Japanese by Joseph Heco [link]


How to Tell a Story
Mark Twain once tried to distinguish between the storyteller’s art and tales that a machine could generate:
To string incongruities and absurdities together in a wandering and sometimes purposeless way, and seem innocently unaware that they are absurdities, is the basis of the American art, if my position is correct. Another feature is the slurring of the point. A third is the dropping of a studied remark apparently without knowing it, as if one were thinking aloud. The fourth and last is the pause.

The pause is an exceedingly important feature in any kind of story, and a frequently recurring feature, too. It is a dainty thing, and delicate, and also uncertain and treacherous; for it must be exactly the right length--no more and no less--or it fails of its purpose and makes trouble. If the pause is too short the impressive point is passed, and [and if too long] the audience have had time to divine that a surprise is intended--and then you can't surprise them, of course.
Mark Twain [source]

tadar via Urban Ticks 11.05.2009

Darpa Wants to Master the Science of Propaganda
[clipped]

Darpa is asking scientists to “take narratives and make them quantitatively analyzable in a rigorous, transparent and repeatable fashion.”

The program is called “Narrative Networks.” By understanding how stories have shaped your mind, the Pentagon hopes to sniff out who has fallen prey to dangerous ideas, a neuroscience researcher involved in the project tells Danger Room.

In the first 18-month phase of the program, the Pentagon wants researchers to study how stories infiltrate social networks and alter our brain circuits.

Once scientists have perfected the science of how stories affect our neurochemistry, they will develop tools to “detect narrative influence.” These tools will enable “prevention of negative behavioral outcomes … and generation of positive behavioral outcomes, such as building trust.”

When the project enters into a second 18-month phase, it’ll use the research gathered to build “optimized prototype technologies in the form of documents, software, hardware and devices”.

In fact, it’s calling for devices that detect the influence of stories in unseen ways. “Efforts that rely solely on standoff/non-invasive/non-detectable sensors are highly encouraged,” [beyond micro-facial feature analysis, dilation of blood vessels/ pupils].


NARRATIVE NETWORKS Oct 07, 2011
Request for proposal, clipped [source]

 DARPA is soliciting innovative research proposals in the areas of (1) quantitative analysis of narratives, (2) understanding the effects narratives have on human psychology and its affiliated neurobiology, and (3) modeling, simulating, and sensing-especially in stand-off modalities.

Motivation: Narratives exert a powerful influence on human thoughts and behavior. They consolidate memory, shape emotions, cue heuristics and biases in judgment, influence in-group/out-group distinctions, and may affect the fundamental contents of personal identity.
Story Diagram

One: Narrative Analysis. Ascertaining exactly what function stories enact, and by what mechanisms they do so… […] This goal serves to ascertain who is telling stories to whom and for what purpose, and to discover latent indicators of the spread and influence of narrative tropes in structures such as social networks, traditional and social media, and in conversation.

1. Develop new, and extend existing narrative theories. …nature of stories, including, but not limited to, a list of necessary and sufficient conditions that help distinguish narrative stimuli from other stimuli. …kinematics and dynamics of story ontology. …structure and function of narratives, including identifying and discussing aspects of narratives that are universal versus aspects that vary considerably across cultural or social contexts.

2. …role of narrative in security contexts. …role and extent stories play in influencing political violence. …political radicalization and how they can influence a person or group’s choice of means (such as indiscriminant violence) to achieve political ends. …influence bystanders-to-conflict in terms of shaping their attitudes and perceptions. …shape the process of negotiation, especially between key stakeholders. …relationship between narratives and the mechanisms that generate and reinforce psychiatric or clinical conditions. …impact of narratives on attitudes and perceptions.

3. …quantitatively analyzable in a rigorous, transparent and repeatable fashion. …establish a framework for the scientific study of the psychological and neurobiological impact of stories on people. …how stories propagate in a system so as to influence behavior. Identify temporal and spatial dimensions of narratives in different media...

Narrative Power: Encounters, Celebrations, Struggles - edited by L. Timmel Duchamp, with an introduction by Eileen Gunn, 2010
Two: Narrative Neurobiology. Since the brain is the proximate cause of our actions, stories have a direct impact on the neurobiological processes of both the senders and receivers of narratives. …explanations of salient narrative psychological phenomena such as engagement, transportation, immersion, and synchronization are highly encouraged.


1. Assay narrative effects on our basic neurochemistry. Determine if narratives uniquely modulate human hormone or neurotransmitter production. […] Determine the manner in which narrative effects change during ontogeny and/or due to socio-economic and other environmental conditions.

2. …impact on the neurobiology of memory, learning and identity. […] Explore the differential influence of stories on neurotransmitter systems as compared to other environmental stimuli. Determine how stories impact the neurobiology of important identity-related judgments, such as whom you consider to be a member of your in-group and whom you count as an out-group member.

3. …neurobiology of emotions. Identify impacted neurobiological emotions…

4. …neurobiology of moral judgment. …neural mechanism or mechanisms by which narratives affect judgments about moral guilt and innocence, or the moral permissibility and/or impermissibility of certain actions.

5. …brain mechanisms related to social cognition. Determine how narratives differentially affect the neurobiological basis of theory of mind and judgments of the mental states of others. Identify and explore how stories influence neural mechanisms responsible for the generation and sustainment of collective action or group behavior. [MTD] Determine if and by what mechanisms stories uniquely synchronize or sustain the neural mechanisms of shared attention, collaboration, joint problem-solving and trust.

Narrative Structure [of Sucker Punch, the movie]
Three: Narrative Models, Simulations and Sensors. …understand others, detect narrative influence, and predict responses. …enable prevention of negative behavioral outcomes… …predict and quantify how and why our behavior changes as a result of narrative interaction.

1. …modeling and simulating influence. Baseline against existing models of the influence process to determine the best extant methods of understanding and forecasting influence. …consideration of narrative-driven mental and neurobiological states and the variables which influence them. …capturing the transition from changes in beliefs, desires and attitudes to actions. [MTD]

2. […] Identify whether existing influence models can be improved by the addition of a narrative “layer.” …ontology and transition states of this layer. Validate and verify the model in at least one potentially iterable testable domain (such as forecasting the success of advertising, movies, public relations campaigns or reception of disaster relief interventions).

3. Develop non-standard and novel sensor suites keyed to the variables and processes identified in new or improved influence models. Baseline against current technologies for detecting and measuring indirect indicators of neural activity (such as capillary dilation, galvanic skin response, eye pupil dilation, gaze direction, micro-facial feature analysis, etc.), and against current standoff technologies for more direct detection and measurement (such as sensing neurobiological compounds). […]

[…]

Racter: Writing Robots
Meme Wars