Problem 2 Flashcards
Little-C creativity
Refers to individual capacities for doing things in a novel way
e.g.: “Everyday creativity”, taking a new route to work, cooking dinner with different ingredients
What does little-c creativity require ?
Complex operations of human consciousness
–> emotional input, affective co-regulation
Creativity
Refer to things that are
a) unique
b) original
c) novel
d) useful
across generations
Creativity research unfolded in 3 distinct waves in the western World.
Name them.
- Focusing on personalities + dispositions
- -> individualist perspective - Cognitive mechanisms
- -> individualist perspective - Focusing on people nested between the domains of knowledge + fields of experts
- -> sociocultural perspective
Why do people assume that machines are or will be better at maximizing positive outcomes ?
Because, AI is programmed to make decisions w/o affective or cognitive restraints
–> bound to precise utilitarian algorithms
Individualist perspective
Suggests that creativity is a new mental combination that is expressed in the world
–> focuses on the person a s a unit of analysis
Name the 3 main assumptions that the individualist perspective holds.
- Creativity must be something new, unique or original
- Creativity involves combinations
- -> derived singular thoughts or behaviors cannot be creative themselves - Creativity must be externalized
- -> made visible to the researcher + public
Sociocultural perspective
Suggests that creativity is the generation of a product that is judged to be
a) novel
b) appropriate
c) useful or valuable
by a suitable knowledgable social group
–> focuses on large scale innovations + innovators that revolutionize entire fields + transcend time (Big-C)
The systems model of creativity tried to answer the question of where creativity exists within a broader sociocultural system.
It did so by proposing 3 interrelated elements that are necessary for creativity to unfold.
Name them.
- Domain of current knowledge
- -> accepted + agreed upon - Individual
- -> alters a component of the domain to do sth novel - Field of experts
- -> decides whether or not the novelty is accepted into domain
In which way is the sociocultural perspective different to the individualist one ?
Here, the individual is not the sole component of the creative process
–> creativity depends on
a) individual talent
b) existing info
c) judgment by experts
Creativity 4.0 model
Refers to a systems model for fourth wave creativity research, that builds on the 3 element of the Big-C and adds AI
–> individual programs AI to perform tasks + process info in specific ways
Big-C creativity
Refers to important large scale innovations + paradigm shifts
Comparing Big vs Little-C creativity.
Which one will matter most to developing computer-based algorithms for AI ? Why ?
Little-C Creativity, because it resembles most normal human thinking patterns
Name the 7 distinct heuristics of creativity.
- Problem finding
- Knowledge acquisition
- Info collection
- Incubation
- Divergent thinking
- Combination
- Convergent thinking
Why are divergent + convergent thinking said to be particularly important for creative output ?
bla
Do highly creative individuals show different brain patterns ?
Yes,
Spontaneous + constructive regions of the brain regulate activity between the default NW + executive NW
–> thus, increased connectivity between the “divergent” + “convergent” regions of the brain result in more creativity
For what reason is it said that machine learning programs could replace up to 50% of the human workforce by the year 2055 ?
Because deep learning programs operate just like humans with interconnected neural NWs that make sense of imputed data
–> make predictions from past experiences + learn through trial + error to become increasingly precise
Programmer bias
Refer to systematic and repeatable errors in a computer system that create unfair outcomes, such as privileging one arbitrary group of users over others
–> programs include the programmers expectations + imaginations on how the world works
AI is designed to reduce our cognitive load by making decisions for us.
But why can’t machine learning one day replace + fulfill more creative endeavors ?
Because, AI is fed info from an existing knowledge base
–> computer program can’t think for its now, because its inputs are purposefully selected
In which way is the domain of the machine world a closed system ?
It only allows enter to experiences that are programmed into it
–> AI is created from the limitations of the human mind, can thus not alter existing paradigms independently
BUT: in contrast the human mind is an evolving organism
Combinational creativity
Using already existing concepts to create something novel
ex.: using LEGO to built something new
BUT: most difficult to integrate into AI
Exploratory creativity
Finding new concepts within the problem space
–> moving thought it + exploring it to find out what there
Transformational creativity
Finding new concepts outside of the problem space
–> space is transformed by altering one or more of its defining dimensions
e.g.: ideas can now be generated that could not have been generated before the change; fundamentally different
Problem space search
Refers to a process in which several states of an instance are considered, with the intention of finding a goal state with a desired property
Psychological novelty
P-creative idea
Refers to things that are new to the person who granted it
–> being creative for our own; similar to little c creativity
Historical novelty
H-creative idea
Refers to thinks that have never occurred in history before, thus novel to everyone
–> similar to big c creativity
Computational creativity
Refers to a branch of AI that studies + exploits the potential of computers to act as autonomous creators
Adaptive creativity
Refers to a kind of creativity that
- delivers surprising solutions to a problem
- changes the way we view the problem itself
Mathematical optimization
Refers to the selection of a best element from a set of available alternatives
Optimization problem
Refers to the problem of finding the best solution from all feasible ones
–> algorithms or heuristics are used that may provide approx. solutions to the problems
Local optimum
Refers to a solution to the optimization problem that is optimal within a neighboring set of other solutions
Global optimum
Refers to a solution to the optimization problem that is optimal amongst all possible solutions
Local search (hill climbing) method
Involves
- starting from an initial configuration
- then repeatedly moving to an improved neighborhood configuration
Heuristics vs Meta-heuristics
- Heuristics
- -> algorithms that do not guarantee to find the perfect solution - Meta-heuristics
- -> higher-level heuristic designed to find/select a heuristic that will find the perfect solution
Simulated annealing
Refers to a meta heuristic that is used to approximate global optimization in a large search space for an optimization problem
–> optimization method, that can be used to solve combinatorial problems
At the heart of simulated annealing is an analogy with thermodynamics.
Explain this.
Annealing allows thermal equilibrium to be attained at each stage
=> at high temperatures, molecules are free to move but when cool, there is minimal energy
Annealing
When some liquids or metals are cooled slowly, the cooling process is called annealing
How is the optimization method of SA analogous to the one used in thermodynamics ?
Because, it is a minimization technique based on nature
–> while it is unlikely to find the optimum solution, it can find a very good solution, even in the presence of noise
Confluence (componential) approach
Suggests that there must be more than one element present in order to produce a creative outcome
e.g.: Systems model of creativity
Default network
Is associated with spontaneous + self-generated thought
e.g.: mind wandering, autobiographical retrieval
–> mid + posterior inferior parietal cortex
THUS: Idea generation
Control network
Is associated with cognitive processes that require externally directed attention
e.g.: WM, task-set switching
–> shows activation when the default NW deactivates
THUS: Selection of ideas by evaluating each degree of efficacy, then modifying them to an optimal standard
Creative process involves a 2-stage process of ___ and ___ .
- Idea generation
- -> bottom up - Idea evaluation
- -> top down
