Anjar Priandoyo

Catatan Setiap Hari

Posts Tagged ‘Artificial Intelligence

Artificial Intelligence

leave a comment »

Artificial Intelligence can not:

  1. Become conscious (as conscious is fallacy, e.g can we create machine that can be tired?)
  2. Predict future (uncertainty that never been happen before)

*Most AI concept is fallacy, things like become omnipotent. Intelligence is basically a flaw, an attempt to have artificial (to make something that, or to perfect something that flaw -which its intention is to flaw is impossible). Language problem. Science is solving linguistic problem.

Do sentient robots exist? Sentient machines may never exist, according to a variation on a leading mathematical model of how our brains create consciousness

Written by Anjar Priandoyo

Maret 17, 2022 at 6:33 am

Ditulis dalam Society

Tagged with

Artificial Intellience is fraud, scam and fallacy

leave a comment »

Why artificial intelligence is fraud, scam and fallacy.

Fraud Anatomy
1.Complex Scheme: AI/ML basically statistics
2.Time Binding Scheme: We are entering the AI era
3.Emotional related Scheme: AI will destroy humanity
4.Voluntary Action

Scam Anatomy
1.Jargon/Made up word: all AI jargon is made up word
2.Contradiction/Juxtaposition/Contrast: AI can replace human (non living replace living)
3.Repetition: Everyday in news, told by everyone

Fallacy Anatomy
1.Penyesatan (equivocation): The future is AI
2.Penyembunyian (concealment): We need to be careful
3.Pelebayan (exaggeration, can be understatement or overstatement):

Scam Anatomy: 
1.Legal registration: Government sponsor this
2.Physical evidence: It is used everywhere
3.Small amount: Learn AI is easy and fast
4.Innocent actor (Fake motive, voluntary action, zombie (computing)

Simple test. Can AI create God. Theist will say no, Atheist will say no.

Quora: Can AI ever become conscious?
Artificial Intelligence: Does Consciousness Matter? (Elisabeth Hildt)

Written by Anjar Priandoyo

Maret 8, 2022 at 7:30 pm

Ditulis dalam Science

Tagged with ,

Jargon Echo Chamber

leave a comment »

It is interesting, all the jargon is scam. I went to the group find some interesting pattern:

  1. Advertisement of Course/Seminar
  2. Advertisement of Vacancy, since 2006 I start to realize Vacancy ads basically scam
    Echo Chamber

Machine Learning Indonesia: 2.5K (page?)
DevOps Indonesia: Aug 2014: 0.9K
Indonesia Blockchain: Nov 2021: 0.7K
Indonesia Internet of Things: 4.7K (Oct 2016)
Artificial Intelligence Indonesia: 7K (Feb 2017)
Taudata Analytics: Data Science, Big Data, IOT (Jan 2014, name changed Feb 2022), 19K
Cloud Computing Indonesia: Apr 2012, name changed Jul 2020: 8.6K

Written by Anjar Priandoyo

Maret 8, 2022 at 5:52 pm

Ditulis dalam Science

Tagged with

Machine Learning in Oil & Gas

leave a comment »

Machine Learning [ref]

Written by Anjar Priandoyo

Maret 8, 2022 at 4:17 pm

Ditulis dalam Science

Tagged with

Emotion

leave a comment »

Emotion basically is our subjective interpretation of our world. Subjective, as it might be interpreted differently (I like sushi, but if I eat it everyday will get bored). Subjective, as it means that it is unreliable, it is biased, for example the measurement of weather which is not accurate (today is hot, but yesterday is hotter). Subjective as it mental states (that can be changed).

Emotion can affect creativity and interaction (Emotional issues is relationship issues). A successful political engagement at its core is a emotion manipulation, where to higher degree a symbol manipulation.

Machine is the best solution for a physical problem (lifting heavy object), machine such as calculator is the best solution for intelectual problem. However emotion could not be replace by machine. As emotion is attribute to human. There is no human that have better emotional inteligence than other human. There is physical and intelectual superiority, but there is no emotion (attribute to individual) or social (attribute to group) superiority.

So how deal with human? wikipedia will not satisfy human need (as Wiki is intellectual). Human will only interested in other human. If emotion is excluded from human, all the human problem will be dissappear. In order to be succedd, human does not need emotion. Although capacity for symbolic thought can be improved by playing music, painting or writing.

Notes:
Manipulation in Close Relationships: Five Personality Factors in Interactional Context
-Surgency (Coercion, Responsibility, Invocation), Desurgency (Debasement),
-Agreeableness (Pleasure Induction), Disagreeableness (Coercion),
-Conscientiousness (Reason),
-Emotional Instability (Regression)
-Intellect-Openness (Reason)

Emotion: mental states brought on by neurophysiological changes, variously associated with thoughts, feelings, behavioural responses, and a degree of pleasure or displeasure. There is currently no scientific consensus on a definition.

Cognitive: the mental action or process of acquiring knowledge and understanding through thought, experience, and the senses (listening, speaking, NLP, sentiment analysis (understanding emotions and feelings), image recognition).

Emotional manipulation occurs when a manipulative person seeks power over someone else and employs dishonest or exploitive strategies to gain it

Ref: Quora: Could artificial intelligence develop emotions, Can AI ever become conscious, Can AI ever replace human creativity in art?

Could AI develop emotions?
Could AI become conscious? no, as AI is not worry about death
At present, the answer to this question ultimately rests upon a certain belief of the person answering it, namely, the belief as to whether or not consciousness is fundamentally physical in nature.

Stage: Physical, Intellectual, Emotional, Social

Theory of Psychological Development
-Perceptual Development: Eleanor Gibson
-Emotional Development: John Bowlby, Mary Ainsworth, Harry Harlow
-Cognitive Development: Jean Piaget
-Moral Development: Lawrence Kohlberg
-Psychosocial Development: Erik Erikson

Cognitive: Piaget’s Preoperational Stage and Symbolic Thought
-Sensori Motor (0-2 years old): Coordination of senses with motor responses, sensory curiosity about the world.
-Preoperational (2-7 years old): Symbolic thinking, use proper syntax and grammar to express concepts.
-Concrete Operational (7-11 years old): Time, space and quantity
-Formal Operational (>11 years old): Theoretical, hypothetical, counterfactual thinking. Abstract logic and reasoning.

Emotion: John Bowlby Attachment Theory
-Preattachment (birth to 6 weeks): sensory prefaces bring infants close to parents
-Attachment in the making (6 weeks to 6-8 months): infants develop stranger anxiety, differentiating from those they know and those they do not.
-Clear-cut attachment (6-8 months to 18 months-2 years): infant develops separation anxiety when a person he is attached to leaves him
-Goal-corrected partnership (18 months on): toddlers create reciprocal relationships with their mothers

Socio-Emotion: Erikson’s Social-Emotional Development Theory (hopes, will, purpose, competence, fidelity, love, care, wisdom)

Written by Anjar Priandoyo

Februari 22, 2022 at 4:46 pm

Ditulis dalam Science

Tagged with ,

Machine Learning

leave a comment »

Written by Anjar Priandoyo

Oktober 2, 2021 at 4:42 am

Ditulis dalam Life

Tagged with ,

Machine Learning and Big Data

leave a comment »

Interesting, some might claim that Machine Learning is part of Big Data.

Written by Anjar Priandoyo

Agustus 22, 2020 at 3:43 pm

Ditulis dalam Science

Tagged with ,

Artificial Intelligence and Complex Jargon

leave a comment »

What is the difference between deep learning and neural network? I heard this question on one of doctoral viva, surprisingly the candidate who learn about this intensively for 5 years, and might be even more, can not answer this question. Well, this is trick question:

  1. DL and NN as historical relationship
    Old age researcher like me, tend to understand things based on the past. In the past, Neural Network is a different kind of “AI” that works only if there is a set of data for training. In the past there is no concept of DL. Therefore researcher tend to think that DL is continuation of NN.
  2. DL and NN might be seen as subset relationship
    ML, DL and NN. The AI research tend to scale up the size of their research. Therefore the biggest domain is the newest. This is a bit confusing, old generation always think that NN is the first, but newer generation think that ML is the first.

E.g convolutional neural network (CNN) is subset of DL, NN

Deep learning is machine learning based on ANN.

Written by Anjar Priandoyo

Agustus 22, 2020 at 6:44 am

Ditulis dalam Science

Tagged with

Artificial Intelligence: Past vs Present

leave a comment »

Past vs Present

The first chapter that I learned in AI is Expert System. I missed the

Knowledge-driven approach vs Data-driven approach
Expert System vs Neural Network
Expert System vs Intelligent Agent

Expert System: Rules and data, Neural Network: Remembering/Learning

Expert System = Inference Engine + Knowlege Base = Information knowledge based systems (IKBS)

expert system tends to imply a particular kind of reasoning (IKBS etc)

intelligent agent
– could mean almost anything that reacts to its environment
– act autonomously on behalf of a user, after acquiring data from their environment.

Expert system implies a significant amount of knowledge engineered into the program. Agent implies a piece of software that (relatively) autonomously interacts with some (fairly complex) environment.

expert system uses sets of rules and data to produce a decision or recommendation. Neural networks, on the other hand, attempt to simulate the human brain by collecting and processing data for the purpose of “remembering” or “learning”

Deep Learning: Machine learning methods based on artificial neural networks with representation learning.

ANN dengan jumlah layer yang banyak disebut deep learning

A perceptron was a form of neural network introduced in 1958 by Frank Rosenblatt,

CNN Convolutional Neural Network: Arsitektur Deep Learning yang mampu mereduksi dimensi pada data tanpa menghilangkan ciri atau fitur

k-nearest neighbors algorithm

Wiki:
Timeline of AI, Timeline of Machine Learning, Progress in AI, History of NLP

Expert System vs Neural Network

Written by Anjar Priandoyo

Juli 27, 2020 at 5:49 am

Ditulis dalam Science

Tagged with

Artificial Intelligence

leave a comment »

As I learn my first AI in 2000s, my understanding of AI is limited to Expert System (Sistem Pakar), which divided into two subsystems: Inference Engine (that think) and Knowledge based (that know based on facts and rules).

Even during the peak of AI in 2000s -which everybody knows that “knowledge acquisition” is the biggest problem, people still hope that by developing an “ontology” which then developed in “semantic web” aka XML, which unfortunatelly again loose by JSON (REST API). In short semantic web -or an attempt to classify is dead, replaced by machine learning / data mining.

In expert system the algorithm is simple: forward chaining, backward chaining and several others such as truth maintenance, hypothetical reasoning, fuzzy logic and ontology classification.

Expert system is basically a probabilistic systems that were plagued by theoretical and practical problems of data acquisition and representation. Expert System is a symbolic AI/knowledge base learning that available in the form of human-readable representations of problems, logic and search. Expert system use network of production rules.

There are several weakness of Expert System, first is the problem of knowledge acquisition the second is on the uncertainty (with lack of exact reasoning, rules will be problematic). E.g if green means watermelon and yellow means banana, however there are some fruit that light green or light yellow. E.g IF sky is clear THEN Sunny we add the forecast is 0.8 (certainty factor)

Machine learning is whole different thing compare with Expert System. Machine learning is

Keyword:
Why semantic web failed
Why expert system failed (slowdown since early 1990s)
AI Spring (circa 2012), The coming of AI Spring (2019)

Written by Anjar Priandoyo

Juli 26, 2020 at 5:09 pm

Ditulis dalam Science

Tagged with