Skip to main content

Home/ GAVNet Collaborative Curation/ Group items tagged reinforcement

Rss Feed Group items tagged

Filter: All | Bookmarks | Topics Simple Middle
Steve Bosserman

Uber has cracked two classic '80s video games by giving an AI algorithm a new type of m... - 0 views

www.technologyreview.com/...algorithm-a-new-type-of-memory

algorithms artificial intelligence machine learning reinforcement

shared by Steve Bosserman on 27 Nov 18 - No Cached
  • AI researchers have typically tried to get around the issues posed by by Montezuma’s Revenge and Pitfall! by instructing reinforcement-learning algorithms to explore randomly at times, while adding rewards for exploration—what’s known as “intrinsic motivation.” But the Uber researchers believe this fails to capture an important aspect of human curiosity. “We hypothesize that a major weakness of current intrinsic motivation algorithms is detachment,” they write. “Wherein the algorithms forget about promising areas they have visited, meaning they do not return to them to see if they lead to new states.”
  • The team’s new family of reinforcement-learning algorithms, dubbed Go-Explore, remember where they have been before, and will return to a particular area or task later on to see if it might help provide better overall results. The researchers also found that adding a little bit of domain knowledge, by having human players highlight interesting or important areas, sped up the algorithms’ learning and progress by a remarkable amount. This is significant because there may be many real-world situations where you would want an algorithm and a person to work together to solve a hard task.
Steve Bosserman

Specifying AI safety problems in simple environments | DeepMind - 0 views

deepmind.com/...specifying-ai-safety-problems

machine learning safety artificial intelligence

shared by Steve Bosserman on 01 Oct 18 - No Cached
  • Steve Bosserman on 01 Oct 18
     
    "As AI systems become more general and more useful in the real world, ensuring they behave safely will become even more important. To date, the majority of technical AI safety research has focused on developing a theoretical understanding about the nature and causes of unsafe behaviour. Our new paper builds on a recent shift towards empirical testing (see Concrete Problems in AI Safety) and introduces a selection of simple reinforcement learning environments designed specifically to measure 'safe behaviours'."
Bill Fulkerson

The Looting Machine Called Capitalism - 0 views

www.counterpunch.org/...ting-machine-called-capitalism

imperialism capitalism

shared by Bill Fulkerson on 27 Apr 17 - No Cached
  • Bill Fulkerson on 27 Apr 17
     
    "I have come to the conclusion that capitalism is successful primarily because it can impose the majority of the costs associated with its economic activities on outside parties and on the environment. In other words, capitalists make profits because their costs are externalized and born by others. In the US, society and the environment have to pick up the tab produced by capitalist activity. In the past when critics raised the question about external costs, that is, costs that are external to the company although produced by the company's activities, economists answered that it was not really a problem, because those harmed by the activity could be compensated for the damages that they suffered. This statement was intended to reinforce the claim that capitalism served the general welfare. However, the extremely primitive nature of American property rights meant that rarely would those suffering harm be compensated. The apologists for capitalism saved the system in the abstract, but not in reality."
Bill Fulkerson

Case Series of Multisystem Inflammatory Syndrome in Adults Associated with SARS-CoV-2 I... - 0 views

www.cdc.gov/...mm6940e1.htm

united states

shared by Bill Fulkerson on 05 Oct 20 - No Cached
  • Bill Fulkerson on 05 Oct 20
     
    Clinical suspicion and indicated SARS-CoV-2 testing, including antibody testing, might be needed to recognize and treat adults with MIS-A. Further research is needed to understand the pathogenesis and long-term effects of this condition. Ultimately, the recognition of MIS-A reinforces the need for prevention efforts to limit spread of SARS-CoV-2. COVID-19
Steve Bosserman

Only governments can safeguard the openness of the internet - Rufus Pollock | Aeon Ideas - 0 views

aeon.co/...d-the-freedoms-of-the-internet

transparency control openness governance internet

shared by Steve Bosserman on 13 Nov 16 - No Cached
  • The internet’s low-cost transmission can just as easily create information empires and robber barons as it can digital democracy and information equality. The growing value of being able to mine and manipulate huge data-sets, to generate predictions about consumers’ behaviour and desires, creates a self-reinforcing spiral of network effects. Data begets more data, locked down behind each company’s walls where their proprietary algorithms can exploit it for profit.
  • Today, the equivalent gesture might be to turn away from private monopolies to fund innovation and creativity. What matters is who owns information, not just the infrastructure by which it is distributed. Digital technology must be combined with concrete actions that protect openness across the spectrum, from maps to medicines, from software to schools. Better that we do it through public institutions, instead of relying on mavericks and martyrs.
Steve Bosserman

The idea of intellectual property is nonsensical and pernicious - Samir Chopra | Aeon E... - 0 views

aeon.co/...-is-nonsensical-and-pernicious

intellectual property

shared by Steve Bosserman on 12 Nov 18 - No Cached
  • A general term is useful only if it subsumes related concepts in such a way that semantic value is added. If our comprehension is not increased by our chosen generalised term, then we shouldn’t use it. A common claim such as ‘they stole my intellectual property’ is singularly uninformative, since the general term ‘intellectual property’ obscures more than it illuminates. If copyright infringement is alleged, we try to identify the copyrightable concrete expression, the nature of the infringement and so on. If patent infringement is alleged, we check another set of conditions (does the ‘new’ invention replicate the design of the older one?), and so on for trademarks (does the offending symbol substantially and misleadingly resemble the protected trademark?) and trade secrets (did the enterprise attempt to keep supposedly protected information secret?) The use of the general term ‘intellectual property’ tells us precisely nothing.
  • Property is a legally constructed, historically contingent, social fact. It is founded on economic and social imperatives to distribute and manage material resources – and, thus, wealth and power. As the preface to a legal textbook puts it, legal systems of property ‘confer benefits and impose burdens’ on owners and nonowners respectively. Law defines property. It circumscribes the conditions under which legal subjects may acquire, and properly use and dispose of their property and that of others. It makes concrete the ‘natural right’ of holding property. Different sets of rules create systems with varying allocations of power for owners and others. Some grants of property rights lock in, preserve and reinforce existing relations of race, class or gender, stratifying society and creating new, entrenched, propertied classes. Law makes property part of our socially constructed reality, reconfigurable if social needs change.
  • ‘Property’ is a legal term with overwhelming emotive, expressive and rhetorical impact. It is regarded as the foundation of a culture and as the foundation of an economic system. It pervades our moral sense, our normative order. It has ideological weight and propaganda value. To use the term ‘intellectual property’ is to partake of property’s expressive impact in an economic and political order constructed by property’s legal rights. It is to suggest that if property is at play, then it can be stolen, and therefore must be protected with the same zeal that the homeowner guards her home against invaders and thieves.
  • ...2 more annotations...
  • What about the common objection that without ‘intellectual property’ the proverbial starving artist would be at the mercy of giant corporations, who have existing market share and first-mover advantage? It is important to disaggregate the necessity and desirability of the protections of the various legal regimes of copyright, patents, trademarks and trade secrets from that of the language of ‘intellectual property’. Current copyright, patent, trade-secret and trademark law do not need to be completely rejected. Their aims are rather more modest: the reconfiguration of legal rules and protections in an economy and culture in which the nature of creative goods and how they are made, used, shared, modified and distributed has changed. Such advocacy is not against, for instance, copyright protections. Indeed, in the domain of free and open-source software, it is copyright law – through the use of artfully configured software licences that do not restrain users in the way that traditional proprietary software licences do – that protects developers and users. And neither do copyright reformers argue that plagiarists be somehow rewarded; they do not advocate that anyone should be able to take a copyrighted work, put their name on it, and sell it.
  • This public domain is ours to draw upon for future use. The granting of temporary leases to various landlords to extract monopoly rent should be recognised for what it is: a limited privilege for our benefit. The use of ‘intellectual property’ is a rhetorical move by one partner in this conversation, the one owning the supposed ‘property right’. There is no need for us to play along, to confuse one kind of property with another or, for that matter, to even consider the latter kind of object any kind of property at all. Doing so will not dismantle the elaborate structures of rules we have built in order to incentivise artistic and scientific work. Rather, it will make it possible for that work to continue.
Steve Bosserman

High score, low pay: why the gig economy loves gamification | Business | The Guardian - 0 views

www.theguardian.com/...ivers-ride-hailing-gig-economy

gamification gig economy taxi service incentive productivity profit

shared by Steve Bosserman on 23 Nov 18 - No Cached
  • Simply defined, gamification is the use of game elements – point-scoring, levels, competition with others, measurable evidence of accomplishment, ratings and rules of play – in non-game contexts. Games deliver an instantaneous, visceral experience of success and reward, and they are increasingly used in the workplace to promote emotional engagement with the work process, to increase workers’ psychological investment in completing otherwise uninspiring tasks, and to influence, or “nudge”, workers’ behaviour.
  • According to Burawoy, production at Allied was deliberately organised by management to encourage workers to play the game. When work took the form of a game, Burawoy observed, something interesting happened: workers’ primary source of conflict was no longer with the boss. Instead, tensions were dispersed between workers (the scheduling man, the truckers, the inspectors), between operators and their machines, and between operators and their own physical limitations (their stamina, precision of movement, focus). The battle to beat the quota also transformed a monotonous, soul-crushing job into an exciting outlet for workers to exercise their creativity, speed and skill. Workers attached notions of status and prestige to their output, and the game presented them with a series of choices throughout the day, affording them a sense of relative autonomy and control. It tapped into a worker’s desire for self-determination and self-expression. Then, it directed that desire towards the production of profit for their employer.
  • Former Google “design ethicist” Tristan Harris has also described how the “pull-to-refresh” mechanism used in most social media feeds mimics the clever architecture of a slot machine: users never know when they are going to experience gratification – a dozen new likes or retweets – but they know that gratification will eventually come. This unpredictability is addictive: behavioural psychologists have long understood that gambling uses variable reinforcement schedules – unpredictable intervals of uncertainty, anticipation and feedback – to condition players into playing just one more round.
  • ...1 more annotation...
  • Gaming the game, Burawoy observed, allowed workers to assert some limited control over the labour process, and to “make out” as a result. In turn, that win had the effect of reproducing the players’ commitment to playing, and their consent to the rules of the game. When players were unsuccessful, their dissatisfaction was directed at the game’s obstacles, not at the capitalist class, which sets the rules. The inbuilt antagonism between the player and the game replaces, in the mind of the worker, the deeper antagonism between boss and worker. Learning how to operate cleverly within the game’s parameters becomes the only imaginable option. And now there is another layer interposed between labour and capital: the algorithm.
Steve Bosserman

I am a data factory (and so are you) - 0 views

www.roughtype.com/?p=8394

data human rights commons platforms

shared by Steve Bosserman on 10 May 18 - No Cached
  • Data is no less a form of common property than oil or soil or copper. We make data together, and we make it meaningful together, but its value is currently captured by the companies that own it. We find ourselves in the position of a colonized country, our resources extracted to fill faraway pockets. Wealth that belongs to the many — wealth that could help feed, educate, house and heal people — is used to enrich the few. The solution is to take up the template of resource nationalism, and nationalize our data reserves.
  • Emphasising time well spent means creating a Facebook that prioritises data-rich personal interactions that Facebook can use to make a more engaging platform. Rather than spending a lot of time doing things that Facebook doesn’t find valuable – such as watching viral videos – you can spend a bit less time, but spend it doing things that Facebook does find valuable. In other words, “time well spent” means Facebook can monetise more efficiently. It can prioritise the intensity of data extraction over its extensiveness. This is a wise business move, disguised as a concession to critics. Shifting to this model not only sidesteps concerns about tech addiction – it also acknowledges certain basic limits to Facebook’s current growth model. There are only so many hours in the day. Facebook can’t keep prioritising total time spent – it has to extract more value from less time.
  • But let’s assume that our vast data collective is secure, well managed, and put to purely democratic ends. The shift of data ownership from the private to the public sector may well succeed in reducing the economic power of Silicon Valley, but what it would also do is reinforce and indeed institutionalize Silicon Valley’s computationalist ideology, with its foundational, Taylorist belief that, at a personal and collective level, humanity can and should be optimized through better programming. The ethos and incentives of constant surveillance would become even more deeply embedded in our lives, as we take on the roles of both the watched and the watcher. Consumer, track thyself! And, even with such a shift in ownership, we’d still confront the fraught issues of design, manipulation, and agency.
Steve Bosserman

How We Made AI As Racist and Sexist As Humans - 0 views

thewalrus.ca/as-racist-and-sexist-as-humans

artificial intelligence algorithms sexism data mining racism

shared by Steve Bosserman on 25 May 18 - No Cached
  • Artificial intelligence may have cracked the code on certain tasks that typically require human smarts, but in order to learn, these algorithms need vast quantities of data that humans have produced. They hoover up that information, rummage around in search of commonalities and correlations, and then offer a classification or prediction (whether that lesion is cancerous, whether you’ll default on your loan) based on the patterns they detect. Yet they’re only as clever as the data they’re trained on, which means that our limitations—our biases, our blind spots, our inattention—become theirs as well.
  • The majority of AI systems used in commercial applications—the ones that mediate our access to services like jobs, credit, and loans— are proprietary, their algorithms and training data kept hidden from public view. That makes it exceptionally difficult for an individual to interrogate the decisions of a machine or to know when an algorithm, trained on historical examples checkered by human bias, is stacked against them. And forget about trying to prove that AI systems may be violating human rights legislation.
  • Data is essential to the operation of an AI system. And the more complicated the system—the more layers in the neural nets, to translate speech or identify faces or calculate the likelihood someone defaults on a loan—the more data must be collected.
  • ...8 more annotations...
  • But not everyone will be equally represented in that data.
  • And sometimes, even when ample data exists, those who build the training sets don’t take deliberate measures to ensure its diversity
  • The power of the system is its “ability to recognize that correlations occur between gender and professions,” says Kathryn Hume. “The downside is that there’s no intentionality behind the system—it’s just math picking up on correlations. It doesn’t know this is a sensitive issue.” There’s a tension between the futuristic and the archaic at play in this technology. AI is evolving much more rapidly than the data it has to work with, so it’s destined not just to reflect and replicate biases but also to prolong and reinforce them.
  • Accordingly, groups that have been the target of systemic discrimination by institutions that include police forces and courts don’t fare any better when judgment is handed over to a machine.
  • A growing field of research, in fact, now looks to apply algorithmic solutions to the problems of algorithmic bias.
  • Still, algorithmic interventions only do so much; addressing bias also demands diversity in the programmers who are training machines in the first place.
  • A growing awareness of algorithmic bias isn’t only a chance to intervene in our approaches to building AI systems. It’s an opportunity to interrogate why the data we’ve created looks like this and what prejudices continue to shape a society that allows these patterns in the data to emerge.
  • Of course, there’s another solution, elegant in its simplicity and fundamentally fair: get better data.
1 - 10 of 10
Showing 20▼ items per page