KISS ME
Tauhid Nur Azhar
Last Saturday at the West Hall of ITB, I received a lot of enlightenment across disciplines. Besides the discussion with Bu Lulu, Prof. Supri, and Bu Renny, I also gained many insights from Prof. Yogi’s presentation, Prof. Marsel’s presentation from the SITH department, and Prof. Miming’s presentation from the SAPPK department.
One of the idioms or acronyms that caught my attention was KISS ME, which was presented by Prof. Yogi on one of his PowerPoint slides. According to Prof. Yogi, KISS ME stands for Keep it, Short, Simple, and MEaningful.
A good mathematical model, just like the solution to various problems in life, should be simple yet not simplistic, concise yet rich in meaning, and not lose relevance with the underlying core values. It should be intelligent, noble, and ideal, and ideally born from sincere intentions.
The concept of KISS ME also emerged from Prof. Marsel’s research series, which focused on cell biology and immunology, and his consistent exploration of the field he was interested in, which has resulted in significant contributions to the development of the field.
Prof. Marsel has extensively studied various physiological behaviors of cells, including pathological or deviant conditions that cause health problems.
He has also studied genes such as GPCR55 and BRD4 related to tumor markers and epigenetic mechanisms in cancer, and has gained knowledge about the function of these genes that can be used in diagnostic and therapeutic processes.
Based on this evidence-based science, research in the same domain can develop, for example, the development of natural materials as a source of active substances in cancer therapy. Prof. Marsel has been involved in this research, such as the use of exosomes from ginger, mangosteen extract, or miRNA.
On the other hand, the development of immunology, which he also masters, has contributed to the development of the Merah Putih ITB vaccine, which was designed using two approaches, one of which used Adenovirus.
Even during the pandemic, Prof. Marsel has contributed to the effort to map the virus genome through the WGS method. The resulting genome map is a valuable asset in constructing diagnostic tools, vaccines, treatment regimens, and data bases in the context of epidemiology.
It has been proven that the contribution of genomic data from various parts of the world, collected through the GISAID initiative, has provided many benefits in the joint effort to eradicate and reduce the impact of the pandemic. The concept of preventing transmission can be equipped with a robust strategy because of the precise genomic data. Similarly, the process of making vaccines and diagnostic tools can proceed because of the objective and scientific data.
Based on the reflection of the learning outcomes from last weekend, I would like to elaborate further on Prof. Yogi’s KISS ME concept, a mathematician modeling expert.
I would also like to propose my own version and interpretation of this concept. The acronym I would like to propose as part of the effort to formulate a smart and meaningful concept is: Keep it Simple, Smart, and MEaningful.
This is a thesis about simple but intelligent and meaningful thinking. Actually, there is one more point that should be included: impactful, having a full impact on a process that constructs truth and goodness. Because what is true is not necessarily good, and what is good is not necessarily true.
This way of thinking is what may be developed together, because it is wandering with attitudes, behaviors, and eventually levels of consciousness.
Colleagues from various disciplines under the umbrella of humaniora often discuss and map out this concept, which is correlated with logic. In my personal opinion, logic as a structural framework (scaffold) can become the foundation for the emergence of exploratory thinking that integrates observation with analytical depth and decision-making processes.
Logical thinking (based on algorithmic logic) in philosophy refers to the use of logic to achieve a valid conclusion based on given premises.
This pattern refers to a structured, systematic, and consistent process that follows certain rules to achieve a correct or most likely correct conclusion.
The characteristics of logical thinking that have been agreed upon so far include consistency, where thinking must be free from contradictions and generally be firm. A conclusion should not contradict previously accepted premises.
Then there is coherence, where arguments or thinking must be connected in a logical and supporting manner, especially to produce validity, where the conclusion produced must follow the correct logical rules. If the premises are true, then the conclusion must also be true logically.
Therefore, clarification is required, where logical thinking requires clear statements and precise definitions to avoid ambiguity, accompanied by a basis and references that can be traced back to their validity.
From various approaches to logical thinking, the method of logical thinking has developed. One of them is deduction, where deduction is a method of thinking where conclusions are drawn from general premises that are considered true. This pattern moves from general to specific.
For example, if there is a premise 1 that says all humans are mortal, and a premise 2 that says Bambang is human, then the conclusion is that Bambang is mortal.
But be careful, because if there is a premise that is similar but potentially contains falsification, then the conclusion may differ based on the dynamic data received. For example, if there is a premise that says all goats eat grass, and a premise 2 that says cows eat grass, it cannot be concluded that cows are goats. This often happens. Efforts to generalize causal relationships in the wrong direction.
Similarly, if there is a premise that says people from Bumiayu are Javanese, it does not mean that all people who live in Bumiayu are Javanese.
Theories related to the method of deductive thinking include Aristotelian logic, which develops syllogisms as a basic deductive structure.
The next method of thinking is induction. Induction is a method of thinking where conclusions are drawn from a number of specific observations to a general conclusion. This pattern moves from specific to general.
For example, observation 1: every time we see the sun rise, it rises from the east. Observation 2: tomorrow the sun will also rise from the east.
The theory related to this method of thinking is David Hume’s theory of induction, which emphasizes that although induction is useful, it does not provide absolute certainty because it is based on habit or pattern repetition. So Hume actually offers an anti-thesis to the inductive approach that is still relative.
The next method of thinking is abduction. Abduction is a method of thinking that tries to find the most plausible explanation or hypothesis based on the available facts. This is often used in contexts where deduction and induction do not provide a direct answer.
Example:
- Observation: The floor is wet.
- Abductive conclusion: M (maybe) there is a leak in the water pipe or it just rained, or there is another phenomenon that is systematically observed and is a valid source of data that can be proven to play a role in causing an event or phenomenon.
Charles Sanders Peirce is a scholar who developed the concept of abduction as a way to generate hypotheses.
Understanding and applying formal logic was developed by Aristotle and later by modern philosophers such as Frege and Russell. In this context, logic involves the study of the correct form of arguments and the rules of logic that govern the relationship between premises and conclusions.
One of the basic principles in Aristotelian logic is that a statement cannot be both true and false at the same time.
In systematic thinking, there is the principle of Occam’s Razor. This principle states that when there are several explanations for a phenomenon, the simplest explanation is usually the most accurate. This principle is used to simplify thinking without neglecting the meaning contained within.
I would like to delve a bit deeper into the principle of Occam’s Razor, as it aligns with the element of simplicity in the KISS ME acronym. This principle is named after William of Ockham, a philosopher and theologian from the 14th century.
In simple terms, this principle states that entities should not be multiplied beyond necessity, or in more common language, the simplest explanation is usually the most accurate.
Occam’s Razor encourages us to choose a theory or explanation that is the simplest and least introduces unnecessary assumptions. In the context of decision-making or problem analysis, this means choosing a solution that does not introduce new factors or variables unless necessary, as long as the solution is sufficient to explain the phenomenon or problem.
For example, when trying to explain why a durian falls from a tree, according to Occam’s Razor, we would be better off accepting the simple explanation of Newton’s law of gravity rather than creating a new theory that is more complex without additional evidence.
In philosophy, Occam’s Razor is used to simplify arguments by eliminating unnecessary assumptions or entities. This can help build stronger and more easily understood arguments.
In science, this principle is used to choose between two different theories that can explain the same data. For example, in scientific theory, a simpler hypothesis is often chosen because it is easier to test and verify.
The main criticism of Occam’s Razor is that simplicity is not the only criterion for truth, or that simple is not necessarily true, so it must be accompanied by smart.
There are situations where a more complex solution is actually more accurate in explaining a particular phenomenon, and what is considered simple can be subjective. For some people, a theory that appears simple may actually require more complex assumptions, and conversely, a complex theory may produce a more accurate solution because it is more comprehensive and can accommodate dynamic phenomena and data.
In the field of psychology, there is the Law of Parsimony (Hukum Kekikiran), similar to Occam’s Razor, where psychologists tend to seek the simplest explanation for human behavior. This is often used in learning and behavior theory to explain simple responses to stimuli without involving more complex mechanisms unless necessary.
There is also the Cognitive Load Theory (Teori Beban Kognitif), which states that the human brain has limited capacity to process information. Therefore, a simpler theory that minimizes cognitive load is easier to understand and remember by the human brain. This aligns with Occam’s Razor in seeking the simplest explanation.
There is also an approach known as heuristics (heuristik). In neuropsychology, heuristics are practical rules or simple approaches used by the brain to make quick decisions. Heuristics often use principles that align with Occam’s Razor, such as choosing an explanation or solution that appears the simplest and most direct, although sometimes it can lead to bias or error.
Actually, in the context of generative AI, as the principle of thinking is replicated by the Transformer model with GPU as its brain, holistic thinking that combines Occam’s Razor with complex dataset and self-directed learning processes becomes a key concept. In other words, there is a leap in systematic thinking that begins with exploring and searching for complex data that intertwines with skill training to utilize it in the process of manufacturing cognitive products, as we can currently see and use.
Most of the main principles used in manufacturing cognitive products that produce smart characteristics refer to the functions of physiological brain and neural networks.
In the context of neuroscience, generating decisions or thinking that aligns with Occam’s Razor or Gen AI involves several parts of the brain responsible for cognitive processes, evaluation, and decision-making.
The manufacturing process of cognitive products, as we have discussed above, biologically involves several functional structures of the brain, including:
- Prefrontal Cortex (PFC), which is involved in decision-making, rational thinking, and executive control. PFC plays a crucial role in evaluating various alternative solutions and selecting the optimal decision that is the simplest, in accordance with the principle of Occam’s Razor.
- Dorsolateral prefrontal cortex (DLPFC), which is involved in decision-making that requires logical evaluation and simplification of complex information.
- Anterior Cingulate Cortex (ACC), which is involved in monitoring cognitive conflicts and errors, and is also involved in evaluating options that may lead to uncertainty or ambiguity. ACC helps to weigh various options and identify the most efficient and simple solution, which aligns with the principle of Occam’s Razor.
- Parietal cortex, particularly the inferior parietal lobule (IPL), which is involved in logical and mathematical processing. IPL helps to integrate sensory information with higher-level cognitive processes to produce a simpler solution, reducing complexity and prioritizing relevant information.
- Basal Ganglia, which is involved in decision-making to make it quick and efficient, especially in the context of choosing between several alternatives. This structure supports the principle of Occam’s Razor by helping the brain to select the most energy- and cognitively efficient solution, thereby speeding up the decision-making process.
- Orbitofrontal Cortex (OFC), which is involved in evaluating the value and reward of various options. OFC plays a role in evaluating the relative benefits of different solutions and often supports a simple and direct solution if it is considered the most beneficial or at least low-risk.
- Insula, which is involved in the context of self-awareness and intuitive evaluation. This function helps to evaluate whether a solution is correct or fits well, especially when a decision must be made quickly.
From the various explanations above, it can be concluded that the manufacturing process of cognitive products cannot be partial or monocultural, it must be produced by integrating various functions at different levels. A relevant example is the step-by-step solution to an arithmetic operation, such as solving the problem 1⁰⁰² — ⁹⁹² = ?.
There are more than one way, of course, to calculate 100 squares and 99 squares and then subtract them, for example. But there is another solution, which may be more systematic, simple, but seems to be represented in a more complex process as follows:
1⁰⁰² — ⁹⁹² = a² — b² = (a + b) (a — b) = (100 + 99) (100–99) = 199 x 1 = 199
A longer process, but each step is clear, logical, easy to understand, and can be learned and replicated. This is the concept of smart.
If this thinking pattern and method can be implemented, then problems can be approached in a simple way with verified and validated data, and processed with logical methodology to produce meaningful knowledge that is correlated with a decision-making system that is assertive, adaptive, and comprehensive. KISS ME!
