Merken
Minds, machines and Python
Automatisierte Medienanalyse
Diese automatischen Videoanalysen setzt das TIB|AV-Portal ein:
Szenenerkennung — Shot Boundary Detection segmentiert das Video anhand von Bildmerkmalen. Ein daraus erzeugtes visuelles Inhaltsverzeichnis gibt einen schnellen Überblick über den Inhalt des Videos und bietet einen zielgenauen Zugriff.
Texterkennung – Intelligent Character Recognition erfasst, indexiert und macht geschriebene Sprache (zum Beispiel Text auf Folien) durchsuchbar.
Spracherkennung – Speech to Text notiert die gesprochene Sprache im Video in Form eines Transkripts, das durchsuchbar ist.
Bilderkennung – Visual Concept Detection indexiert das Bewegtbild mit fachspezifischen und fächerübergreifenden visuellen Konzepten (zum Beispiel Landschaft, Fassadendetail, technische Zeichnung, Computeranimation oder Vorlesung).
Verschlagwortung – Named Entity Recognition beschreibt die einzelnen Videosegmente mit semantisch verknüpften Sachbegriffen. Synonyme oder Unterbegriffe von eingegebenen Suchbegriffen können dadurch automatisch mitgesucht werden, was die Treffermenge erweitert.
Erkannte Entitäten
Sprachtranskript
00:00
few and thank you very much so quickly and introduce myself and on the new structure that's me and my family portrait of me as as I am Python software and all the action I
00:23
work at the Swiss Django companies and and underdeveloped of Django CMS and find the call another of the Django project and a member of the Django Software Foundation board and you can find in the usual place online on as evil linking on IRC and get on twitter and this might even in the process of feel free to contact me about I think so that's the thank you and thank you for inviting me to be here for being here to listen to me speak and thank you especially to the organizers and volunteers of this conference because if you're enjoying yourself it's because of then that I know how much work it is with conference and to the conference of 1 and a half thousand or so people is a pretty big deal things will go wrong if things go wrong is not because they didn't care didn't think it's just because stuff happens so when if you see any of the volunteers organizers just say to them personal thanks by the women during the conference because I can tell you somebody who has done this that actually it means a great deal of time so thank you to the the yeah yeah and got closer to the sponsors who make it happen make possible financially the speakers who have put a lot of time in and all that participants have come here to get together as a said I'm at work on gender CMS was sponsoring so when and we are also very busy this week we got gender and desk on Thursday uh we're doing adopt a workshop on a friday and I was watching so and I believe all excited term that this is a great marketing opportunity and the best that came up with this gender CMS we're around your neck so and I don't know maybe Twitter will come with the that's the I'm going to go into the office next to get also congratulations from the team in that and make something left be the thing we around it actually works really well that's not think so I will
02:36
talk about mines machines and uh Python hour ahead of tiny amount of like encoding and it's and this is Alan Turing
02:48
degree and during the directed graph of him smiling and in his paper Com puting Machinery and Intelligence in 1950 he turned the question of whether machines can think onto its head that that the question Can machines
03:04
think is to meaningless to this discussion and instead of asking whether both can occur inside a computer he wanted us to consider whether instead with the machine could in principle the indistinguishable from a human in those respects that allow us to save human thinks so has intelligence they tend to look at the question about machine intelligence intelligence also upside down and he wasn't at applying insights from philosophy of mind that emerged in the early 20th century which also moved away uh move the questions of mine intelligence away from a concern the metaphysics of thought and intrinsic nature of consciousness towards questions like what constitutes an interaction with intelligence for how do we recognize our minds so in other words he was a very much a
04:01
manual for input that that's the right word of thinkers like Gilbert Ryle or big can argue the search for not a search into a mysteries but a matter of recognizing what was in
04:16
front of us all along so we consider this represents a kind of
04:21
fall in the study of artificial intelligence and Chileans argument wage freed researchers from having to climate and these kind of the mysteries of what goes on inside of consciousness thought um and instead concentrate on producing for example interactions that seem like encounters with intelligence and that's been the dominant force if the light in office of intelligence since then and it's it's up all around us I mean literally our pockets so you put your your phone you can say you do I need to carry an umbrella today or you know what's what's happening what events are running in Bilbao about this week and the power of machines to recognize stuff like languages faces text road traffic or not during mitosis of and and to respond appropriately has taken me by surprise and and I think is taken many people by surprise just how good machines have become of doing this and by terms account I really do think that we're living at the beginning of an age of machine intelligence but what about that other people that new media FROM which fewer people have worked on and we hear less about but people have been working there and there have been trying to unpack these mysteries of thought itself so Turing was a very intelligent think about just because he dismissed something has to meaningless to deserve discussion doesn't necessarily mean that that is so so his you to the people who thought otherwise disagreed with oranges inbound and in In computer intelligence was that she was the author of analyzes that the program analyzer um which was his attempt to demonstrate what in a concrete way what was unsatisfactory about
06:21
Turing's this or the philosopher and John so so in this talk I want to go back to Back Fork because I
06:32
also think that Turing's analysis however the brilliant and powerful it it was is a bit broad and inadequate and I think that the incredible advances all of you this series in the chat bots and self-driving cars as far as we are actually concerned with intelligence I think those dead end so those things like like joint approach begin from the outside but not really I think concerns the nature of intelligence itself but with the challenge of creating an appearance of that what this intelligence look like what is it you to be in the presence of intelligence and narratives fears disappears can be extremely successful and more successful their appearance the more easily when we fall into using the language of intelligence around that behavior sometimes forgetting that I mean something might actually find to that I think that we dealing there with the simulacra of intelligence it's really in the case of C Series no terms of how it is known closer to consciousness that study and I think the more interesting question is the neglected 1 that begin from the inside the asks what lies at the heart of intelligence Howard weight of consciousness arises in as programmers I think we have a concepts and tools to investigate this in useful and interesting ways the 2 things in this talk firstly the programming concepts us the most that the programming concepts that give us the most interesting insights on that most basic programming concepts and that we can find their counterparts in the work of artists poets and writers and that this helps us understand so for this will turn to the concept of course it's a Greek word uh it's would would poetry come from and it means to me or to produce and another thing making more all production the sense of
08:41
manufacture it think all of you French retirement words like federal firing at which they needed the same time make and do all thinking English about where we would say things like use making friends or making a lot you know that that's the kind of making perhaps that has emerged that involved in places prices is not concerned with the material technical construction but with rendering a transformation in the world some and act through which a process through which something new is brought forth and implicit that something be kind becomes another kind of thing all together and you think emerges out something something that wasn't there before so quality can be considered a kind of choices and I think that programming also represents courses and that permits and programmers because they work in for this this this kind of making can help us understand usefully some quite deep questions about ourselves and to about the nature of our human consciousness of course on the mind so this go goes straight into the work of programmers and
09:57
and programmers seem to be particularly fascinated by rule rule-governed rule-governed play when they find it and they respond to strongly especially and in poetry and art for example and there are some poets and artists and writers whose work idea so the very strongly to programs because of its and I think that there's something in program or no I think this is in a way that programmers things that make them especially ready to understand and appreciate the intersection of these things these work and ideas so there's things you know when you the program is just going to get certain ideas for certain work or something like that I don't have the answer is that rule rule-governed play takes place in systems which of course program also a very fascinated by and for other things that programmers not best systems are interesting outside programming self-reference and hierarchies and these also the things that make rule-governed play very interesting and of course there are also some of the most basic structures and concepts in programming cell wants to determine the weights programming works so that subjects even the simplest we're represents however the matter how trivial it is you can still unleash something infinite on the computer will try to realize so the rate constructs in computing this kind of most basic but also the most beautiful and powerful and a lot of of all of parts still the fascination of and go to 10 is quite limited and the only thing interesting about that particularly is that it's infinite not the interviewer comes out of so when that's not very interesting that when we apply leads to other things that make get more interesting does anybody know will remember their website textile doll and I spent a long time that 15 years or so ago what it does it loops over text and represents a visually I can't show it to you on a real website because it it's all there it's Java and and I have to know the virtual machine running Internet Explorer do that so this have shown this is the text of Alice's Adventures in Wonderland right around the edge and then words in the middle in that case depending on where and how many times they appear on the and so look his word and therefore the character alice in relation to the entire text or in relation to the story Alice is literally at the center of the story of Alice's Adventures in Wonderland is the Griffin as you can see a peripheral character is a linear sequence through the actual text in text ideas and go and find an
13:00
it's wonderful if you can get it working on your machine you can find it working on a map of uh Macintosh tell me how does it look to get it working again and so intake start we no longer have a sense of the content of the but instead we get its structure exposed by being presented as metadata and structural
13:22
metadata matter we lost something the content but we gained something else because it lets us see new meaning that perhaps was previously previously obscured and it in the loop of text article see who is associated with whom or what we can see who appears when and with whom we can see dominates the story and when I say see we literally visually can see so don't be distracted that you take stock is nice is amusing literally
13:58
what it does is powerful and significant in fact that's not lost on large corporations and governments to understand the power of this stuff to be on Facebook know all about the structure and metadata and right now in various
14:10
countries governments are anxious to have legislation enacted that will help them look more closely at meta data of course they never do anything like eavesdrop on its citizens because that would be wrong they just want to know that you know so that metadata sounding those conversations about that if you can see that from there storing
14:28
metadata is not an invasion of privacy is a tool that keeps America safe no 1 is listening to your calls congressmen I see here last Friday and torture wife immediately called next this then hotel of course we don't know the content of this calls to know that you want to know never so this stuff is a powerful Microsoft like that came across this by chance that we use the cloud to visualize information so we can track the criminal you have that you think so on takes up the 1st text you get is some Hamlet smoke
15:08
there by coincidence programmers and people doing stuff didn't and always go far from the outside uh and it's all Hamlet these the 2 texts they always go for this means that about these 2 texts his hand and
15:25
his is carry on his own he was a Mexican artist and going to play you a part of his and Hamlet with 2 voices that could have been missed down and we don't have direct orders of the students and number dancers go venerable France's novel Francisco Fernandez Francisco novel Francisco novel France's is going arrays you must then Francisco myself Francisco sentences or not OK so that I think it is I think this is hilarious that funny but I think
16:02
it's not just on it's a new interpretation of Hamlet and to voices left and right is the whole time at a loss about 15 minutes like that we out any content what so ever all we have is the structure so what we lost and what are we left with well it's true that you would know from this that much is about revenge on the goal size and wouldn't and hundreds a principal source and then but this a lot to learn you know is the beginning we know who's there at end which is unique about the end a Shakespeare tragedy then we know that that's a big deal you know everybody gets to be at the end we can discover whom the story revolves around who dominates the conversation hangs around who who answers so we've lost something we deliberately from some information away but we exposed perhaps some previously hidden information at the same time you know when I played this people and person what what is this and then it divides the world up into programmers and non-programmers there was no point no program what's the point and that baffled or or even irritated and I've had people really annoyed by that and the problem solving now you may only program is if you didn't love back then MIT maybe just the soul of the a program of so because they also some people who are not programmers love that's me that non-programmers we have the soul of the anyway the point is that programmers get this kind of thing they understand loops they understand what's important about metadata and they know how to recognize a key value pair
17:46
when they see 1 so you know you should some key value pairs and watching some could actually so that we quickly jumped into um he has set up a little program Python program OK
18:00
and just jump on the Internet users beautiful see this is a very useful library for exploring uh the structure text like HTML it loops through the text try things out and in pairs and then pass them to the same command so I'm going to work this is my little tribute to win the playoffs so but not Franciscan but not a Franciscan but not consistent not consistent but not Franciscan but not of Franciscan and the ratio of non-synonymous so Cisco and nonsense said thank you from just about any so computers
18:54
are perfect tools for uh this kind of
18:57
thing for the exploration analysis discovery of texts have language and if you're interested in doing that it's very easy and accessible stuff to do what could you discovered a simple program you like pick samples of magnets and could you extract the punctuation from the text maybe and recognized or to by the punctuation or by the white space and spend more time with the Raspberry Pi and little decay is an array of NTDs visualizing the characters and Hamlet is kind of spiraling out would have different colored light so that it is from the unit you're allows programmer like this a massive amount of of this stuff going on in the world of literature there's a whole field of experimental literature that
19:45
analyzes text size then deconstructs and this thing we call the new while others through until the workshop of potential literature French mostly French writers mathematicians and artists
19:59
and then we have time talk too much about that I recommend having a look at the back wall and you have to go to experiment literature here a
20:08
couple of pages from the the Cubans Microsoft's thousand 1 page and the consonants on the other hand we can play endless games on games illuminating gains text like this and I'm talking about games and playing and some of it's fun but it's also serious because played doesn't just referred to as having fun planes looseness in medical mechanical connections has a lot of play in this connection when connections not tied to the extent to which 1 side side of a connection is free and I'm determined by the other and plays and possible when you have a linkage where you have a connection of constraint if there were no constraints told them that would be nothing but would you wouldn't have any place just have no connection and on the other hand if the connection was so tight that they will look solid there would be no place either so you may find it with is a connection and where the joint amid some movements and the rules and connections in language just open and not uh open and this enough for both rigorous meaning and play within the same system which is
21:22
quite remarkable and this kind of place something that's enjoying exploited by artists and writers and programmers so we see some examples of grouping in text documents and and how much this kind of play becomes even more interesting when the looping becomes self referential so his carry on again this 1st
21:42
Spanish lessons and some of you already know Spanish but that let me tell you his 1st Spanish lessons so this is the English it's annually this by your this is by this is a neuron C this is new this this is by new see if this is a this is faced by neuron is a neuron see if this but you this is if if it is unknown is unknown see if this is by means this is by if CSS but yours is by see see if this own yours as is a new if the essays by neuron is a neuron C C S is binary this is by no it's if this value on lot CS unless it's by no this is the noise is along this but you see this in the back of head so I and this is language about language this is the language
22:50
looping over its structures recursive on itself it's language teaching language and nothing and again you have this to people and and watch people get angry and angrier as this goes on you know differently in In not people with the souls of program and I have 1 person's total me of music is the shift way to learn Spanish and a lot look up this is carry on the the of this is so fascinating usually when something consumers itself in this way it reduces itself to nothingness but in this case something new something other seems to be coming out something new an almost magical you know something for a maddening about it and it's a perfect example of playing with rules and processes so as a programmer uh you name it to each other in a way this is a program that follows its own internal logic of bases and rules an amazing thing you could make 1st Python lesson
24:02
in In the same way and as you know we have the technology um how we could get Python to consume itself and regenerate itself like that but I got distracted by my Raspberry Pi McCullough Davies incentive already got much much further than that you maybe some user uh uh a better programmer analyst less easily distracted could come talk about so it's any kind of impossible magic when you make something each itself but it's a very old and very ancient concept so here's the real world the snake eating its own tail and it's very important it this it's chemists and their dreams as elective studied chemistry so
24:47
obvious to the coolest had had a dream of the snake eating its tail and providing with the solution to the puzzle of that the benzene molecule after many years of studying carbon carbon uh bonding is an ancient symbol of renewal and regeneration in many cultures and programmers as much as artists of fascinated by what happens when a cell self reflecting process loops and as programmers with particularly lucky because people are perfect tools with which to explore and I was mentioned magic has a kind of impossible magic magic is not just a snake bites its own tail because you know immediately dog bites entailing and not producing anything magical the magic alive when something eats itself and then somehow sustains itself and then something new the merger something that wasn't there before and I think it happens in the source code or something new comes out of loop that wasn't in it when we started a new thing altogether and this is the voices that we're in search of the I'm in search of here's another representation of the robots this time that the artist and the action was also
26:03
fascinated by an instance of reference the wood block print some very interesting things going on at where this is the beginning and i is restricted to the well is it 1 we present several routes was a full bands going to make that 5 loops is falling to that 1 an it's looping on different levels you have to choose the level at which you see a looping left to make a decision to hold in your mind you but
26:35
this hierarchy of looking going on and this is where things
26:40
get really interesting here is that the source of the revenue is possibly the most 1 most interesting thinkers you would ever hope to read a book right he's interested in the hierarchy in multiple kind hierarchies of self-referential beats his best-known book is going to mention about which explores are the music uh logic and consciousness and its own part the model driven by this
27:08
fascination with and so reflecting processing and he's in search of the ultimate in self-reflection which is our own human and self-consciousness subjectivity itself and the book itself loops and recurses reflects itself at different levels in its own structure and he's interested in what could emerge from this kind of process and it's a work of or use humor and sense of esthetics will appeal very naturally to programmers useful model I always think of
27:45
you is that even when you take into account Hofstadter's Law message is not have them so we talked about emergence with something new and unexpected comes after all that process at some other level possibly interested in these emergent properties of systems an emergent property is 1 that arises in the system but can't be found in its components so uh sometimes the problem might be geometrical order this is of rocks in the river where water erosion has produced this model effective brought some shaped like that has some structure the the rocks it's and until there's nothing in the rocks that represents those that structure that's only an interaction between the rocks and the water and the water distance the rocks and rocks
28:43
intending something to the water so the rocks themselves and have structure another example might be and and column where 1 organization is nowhere to be found in the behavior
28:54
of individual and you know you don't see and doing that you come predictive and yet a column a new system emerges that didn't exist before or in part in the work of Richard Wright rest of the procedure she follows a line down the page and it produces through the natural for free play this activity something that's not in any of its components or google deep dream that uses iterated pattern searching and generation of algorithms and matches parts to wholes in courses levels of hierarchy and new levels of significance emerge and evolución natural selection uh in nature itself can be considered an emergent property something that comes out of the basic chemical or biological processes and there are multiple levels of hierarchically systems other properties imagine themselves other emergent features for evolutionary so this property that came out in turn feeds back into the system so there newest behaviors in nature whose explanations of being sold in emergency like cooperative behavior which is hard to explain sometimes wise the cooperation within between species when it would seem that are applicable in individuals has no such benefits and so this graph is from a Python library Axelrod and it shows the successes over time of different strategies in
30:33
atonement of the iterated prisoner's dilemma which was set up as a search for the secret of cooperation in evolutionary 1st by uh Robert Axelrod was a political scientist man between 80 and in the 1st column itself the strategies that what you know the present at time to explain the prisoner's dilemma by the we most programmers are well look up the prisoners dilemma and so strategies did poorly and last year of my friend is a light at Cardiff University implemented this is a python library and its you can you can play you can develop your own strategy in and put it into the torment and it's been wildly
31:15
successful and has reignited interesting in actual world and the and the prisoner's dilemma very interesting so I have a look the actual so these are
31:29
examples of all the emergent properties of different kinds not all of them have anything to do with consciousness of course but
31:36
hostile argues that consciousness intelligent is an emergent property that arises from the systems in the brain and we discuss the use of of all the magic seems the following become self-referential and something new screens after them and then I have sort of interesting hierarchies so we have loops within groups and groups at different levels in the system and his thesis is that hierarchies of self reflecting or self similar to that repeat themselves at different levels lie at the heart of cognition and consciousness the brains and neurological processes are themselves based on the use of self reference and logic and consciousness says it is an inverted property it doesn't exist in the neurons of the brain you know if your intelligence it's not because you have a really intelligent neurons there you will find intelligence of consciousness by looking in to the brain and looking more closely at its neurons if we're going to find something there will be by studying the system what happens in and 2 and between so this record of here we are we're the this is the again
33:09
this is is the impossible magic square something emerges reflection itself feeds back into itself crossing these levels all hierarchy Hofstadter thinks it's obvious that this is the origin of consciousness that self referring characters and a layer secret for the cognition In these nearly
33:37
physical processes in which actually no consciousness reside because in the physical processes consciousness belongs at a different dimension so could consciousness be system that nourishes itself like a snake tedious tail apparently impossibly um and that sort of thing so in fact 1 of his later books is called I Am a Strange Loop and the all our consciousness of subjectivity is this process of grouping and self-repairing across hierarchies so now we're talking about intelligence
34:17
and programming in a very different way of the children talking weight in ways I think to make it possible to ask meaningful questions about where whether machines could think and it raises the possibility of an approach into cognition research and offers artificial intelligence that's quite unlike some of the ones we see at the moment the government be on big Data a huge ontologies and uh reports on kind of on on your mom own lost computing power and instead we begin with the simplest of things like the simplest concepts and tools that we have programmers uh uh enjoy playing with play itself we enjoyed playing with payment role in policies with analogy-making with processes so this approach
35:11
comes into the question of consciousness of intelligence as if it as a process of that is at the same time transformative that brings or something you out of the world War into the world and it is very intellectually compelling and I think at the same time however but also philosophically profound it might be the opportunities for exploring in all right to handle even the most novice strengthened program which is also misleading uh things and it gives us a way into this
35:57
quest that seems very rich and valuable and eliminating rich forever which were exploration especially by outside programmers but it's more than that because I think it has something even more important it gives us a way to look for not just uttered any insight into artificial intelligence but just into intelligence itself and I think the kind of
36:20
post shoring of doing this doesn't give us any of those insights and no I don't think that Microsoft you might take the Microsoft here running loving sex maniacs chatterbot on Twitter that I don't think that tells us anything about the nature of intelligence how ever clever it might b and the question of whether hostile that in the end is is right or wrong I don't think you right wrong and is that important it doesn't matter what is on the right track about nature of consciousness and cognition and his ideas of so beautiful and
36:58
powerful and compelling that we should rest with them anywhere in and certainly more elegant than uh table serial and all these other things that people are doing very clever things with and of about my background signal in programming but in philosophy have been wrestling with cost these ideas for 1 of the little time and and it's it's really wonderful there is the light unhappy with the delays and unable to be in the company of
37:34
programmers who find this stuff by again it was about right souls they find this stuff so fascinating immediately and and also find also fascinated by the poets and writers and artists that me and these things that the I found in programming In a way it come back in a circle for me because there is not a pleasing circle something going back to the beginning these philosophical questions that have been with me for decades something is like a snake eating its tail again and there are some references and I believe you want to
38:13
work full of any of these things that will make the most of southern that bag it's it's just falling apart and it's been with me for a long time and very grateful to to that of other uh so I read that I haven't really taken and but I something to think about it and I have some people come up to me is a better programmer and and give me some better code that I couldn't maybe development in some of
38:44
people like and and let's talk talking to and thank you very much that is what interview questions and we again thank you fuel and I should say uh a colleague of mine who actually was able to come so doing another talk in
39:08
case you get sick of hearing the already on the table documentation driven development so that the city was requested that happens to what you
39:15
know all all later the and so we have I heard some critical like this so a big data approach storm and building intelligent systems and perhaps you can say that they you cannot really call them palaeontologist moment out but imagine they they succeeded and they made their and they made like talking and thinking Robert that you can't have a conversation with our I guess you would still say that that's just a simulation of intelligence so I wondering if you what is the difference between the real intelligence in the simulated intelligence and the will he also there is right we could have an artificial intelligence and I'm not sure that he is right but I think much more head to admit or a host of the light in artificial intelligence the 1 that is based on you know this as you said in the brute force big data mining and machine learning because I don't think it's simulation is good enough so this artificial is different from something that's a simulation just because something often use artificial doesn't mean it's not really that just means that is being made and so we have a real intelligent but real intelligence that's artificial about a real intelligent intelligence simulated itself conviction in terms of its immunostimulation however good it gets because what's going on in the question what's going on inside actually does matter at the personal yeah well this is that this is during cellular it just seems unsatisfactory to me I don't care what's inside of the negotiation 1 could take the value of our a more recursive find view about that that these uh we we saw uh and the Spanish lessons the Spanish lessons basically uh points to the radio of the Our any history of the Spanish is that the Spanish you get what I mean but the thing is that you get many of the vector that concept exists linguistically still there is infinite is not is like variance there and they fix it find of such such sequence there's so we for language we we can we can even represent that sort of thing fix finds of infinite sank which sequences that sort of thing sometimes and bond exist so we we had seen the fires we've got without C 5 and that thing that that is itself maybe something with without the C 5 also the you know the 1st Spanish lesson I'm afraid is a bit of a trick because it's not just holding up entirely by itself is actually the director of the artist who makes it go in the right direction is is not simply putting this is by no at the beginning of uh everything is choosing the right things and it's it's playing with them for example the fact that some things on the Internet that you know the same sounds can be rendered differently so it can keep asking questions so it would actually be an less interesting for wind direction so there is some energy going into that system from outside and that I think is 1 if there's something wrong with of that that's probably where it lies end of the you know damn time formulation along and the talk of do you think that maybe neural nets combine the 2 approaches sort efficient bonuses in neural now have married this emergent feature to them which model to describe those conscience I mean that that seems to be very promising and interesting because they have 1 of the things that we've learned from your your networks is that they sometimes do unexpected things and not just unexpected things that it was elected sort of right but who knew orders all things sometimes when a neural network will have an answer have a response that seems to have stepped out and been looking at the thing from different level so that might be a very uh I didn't know enough about neural networks actually to same thing that useful but what understood is that actually that is a promising possible my over all the data that the of we we that you know or that that is the kind of thing or a thing that might just be the start of the kind of of thing will find more surprising things still unsure what take more questions offline for the 2 trillion years thinking that you know
00:00
Umsetzung <Informatik>
Einfügungsdämpfung
Vektorpotenzial
Spiegelung <Mathematik>
Momentenproblem
t-Test
Mögliche Welt
Selbstbezüglichkeit
Gesetz <Physik>
Statistische Hypothese
Computeranimation
Richtung
Homepage
Chatbot
Metadaten
Algorithmus
Gefangenendilemma
Gruppe <Mathematik>
Mustersprache
Gerade
Auswahlaxiom
Metropolitan area network
Array <Informatik>
Verschiebungsoperator
Kraftfahrzeugmechatroniker
Schlüsselverwaltung
Kategorie <Mathematik>
Gebäude <Mathematik>
Computervirus
Bitrate
Biprodukt
Ereignishorizont
Entscheidungstheorie
Gefangenendilemma
Informationsverarbeitung
Forcing
Rechter Winkel
Programmbibliothek
Ordnung <Mathematik>
Tabelle <Informatik>
Instantiierung
Subtraktion
Folge <Mathematik>
Facebook
Zustandsmaschine
Selbst organisierendes System
Wasserdampftafel
Geräusch
Mathematische Logik
Whiteboard
Virtuelle Maschine
Loop
Weg <Topologie>
Informationsmodellierung
Spieltheorie
Endogene Variable
Canadian Mathematical Society
Programmbibliothek
Gruppoid
Abstand
Inhalt <Mathematik>
Algorithmische Lerntheorie
Datenstruktur
Maßerweiterung
Ganze Funktion
Varianz
ART-Netz
Analysis
Datenmodell
Symboltabelle
Schlussregel
Unendlichkeit
Turing-Test
Simulation
Wort <Informatik>
Steuerwerk
Streuungsdiagramm
Neuronales Netz
Bit
Programmiergerät
Prozess <Physik>
Punkt
Natürliche Zahl
Selbstrepräsentation
Formale Sprache
Gruppenkeim
Snake <Bildverarbeitung>
Familie <Mathematik>
Eins
Übergang
Internetworking
Mooresches Gesetz
Erneuerungstheorie
Einheit <Mathematik>
Schwebung
Stützpunkt <Mathematik>
Flächeninhalt
Strebe
LES
Konstruktor <Informatik>
Parametersystem
Interpretierer
Prozess <Informatik>
Viereck
Reihe
Systemaufruf
p-Block
Quellcode
Ein-Ausgabe
Algorithmische Programmiersprache
Gesetz <Physik>
Arithmetisches Mittel
Soft Computing
Datenfeld
Künstliche Intelligenz
Datenstruktur
Twitter <Softwareplattform>
Geschlecht <Mathematik>
Programmanalyse
Mathematikerin
Strategisches Spiel
Projektive Ebene
Information
Schlüsselverwaltung
Message-Passing
Geflecht <Mathematik>
Sichtbarkeitsverfahren
Ebene
Nebenbedingung
Maschinencode
Web Site
Gewicht <Mathematik>
Turnier <Mathematik>
Metadaten
Hausdorff-Dimension
Physikalismus
Gruppenoperation
Interaktives Fernsehen
Hierarchische Struktur
Zellularer Automat
Zahlenbereich
Transformation <Mathematik>
Kombinatorische Gruppentheorie
Term
Data Mining
Datensatz
Multiplikation
Software
Stichprobenumfang
COM
Zusammenhängender Graph
Automat
Softwareentwickler
Optimierung
Widerspruchsfreiheit
Touchscreen
Schreib-Lese-Kopf
Autorisierung
Beobachtungsstudie
Datenmissbrauch
Kreisfläche
Ontologie <Wissensverarbeitung>
Graph
Zeitabhängigkeit
Kegelschnitt
Relativitätstheorie
Magisches Quadrat
Routing
Vektorraum
Physikalisches System
Hochdruck
Quick-Sort
Schlussregel
Roboter
Office-Paket
Mapping <Computergraphik>
Energiedichte
Formale Sprache
Hypermedia
Mereologie
Serielle Schnittstelle
Innerer Punkt
Term
Verkehrsinformation
Metadaten
Formale Metadaten
| Titel | Minds, machines and Python |
| Serientitel | EuroPython 2016 |
| Teil | 29 |
| Anzahl der Teile | 169 |
| Autor |
Procida, Daniele
|
| Lizenz |
CC-Namensnennung - keine kommerzielle Nutzung - Weitergabe unter gleichen Bedingungen 3.0 Unported: Sie dürfen das Werk bzw. den Inhalt zu jedem legalen und nicht-kommerziellen Zweck nutzen, verändern und in unveränderter oder veränderter Form vervielfältigen, verbreiten und öffentlich zugänglich machen, sofern Sie den Namen des Autors/Rechteinhabers in der von ihm festgelegten Weise nennen und das Werk bzw. diesen Inhalt auch in veränderter Form nur unter den Bedingungen dieser Lizenz weitergeben |
| DOI | 10.5446/21122 |
| Herausgeber | EuroPython |
| Erscheinungsjahr | 2016 |
| Sprache | Englisch |
Inhaltliche Metadaten
| Fachgebiet | Informatik |
| Abstract | Daniele Procida - Minds, machines and Python Are we looking in the wrong direction for artificial intelligence and machine learning? I'll discuss an older but perhaps more satisfying approach, that has been neglected in recent years. It begins with questions in logic and language, and can be explored using easy techniques. I'll use simple Python programs to explore three key notions in this AI research: **loops**, **self-reference** and **tangled hierarchies**, themselves directly reflected in important programming concepts. ----- In recent years, we've seen interesting and spectacular successes in artificial intelligence and machine learning, made possible by leaps in computing power and techniques able to harvest vast quantities of data. The results are uncanny. We see them everywhere, from the personal assistants built into smartphones to the neural networks that do an astounding job of recognising images. However, they're also susceptible to the criticism that they represent not intelligence but a mere simulation of it, and that producing a convincing simulacrum has become more important than a genuine search for intelligence or learning. At the same time, another, perhaps deeper, approach has become neglected in recent decades, along with the questions it asks about the nature of mind, intelligence and learning. This approach begins with fundamental questions in logic and language, and can be explored using some of the simplest programming techniques. In this talk, I'll use simple Python programs to explore three key notions in this strand of artificial intelligence research: *loops*, *self-reference* and *tangled hierarchies*. The way these concepts directly reflect important concepts in programming suggests that for the programmer, this approach could be more interesting and satisfying, and simply more **fun,** than using huge ontologies and big data to create mere simulacra of intelligence. The examples I use will be concrete and easy to understand, even for novice programmers. |
Ähnliche Filme
