Kant pursues this project through the first two chapters of the Groundwork.
Footnotes Epigraphs "if you confront the universe with good intentions in your heart, it will reflect that and reward your intent. You may have a heart of gold -- but so does a hard-boiled egg. In these instances, consequentialists look like their moral evaluations are based on intentions.
Some of the other stories told about him were that he never learned to swim and Are good intentions necessary for moral glasses," she wrote.
The current was very strong and the water ice cold, flowing as it did straight from the mountains. Luckily for young Adolf, the son of the owner of the house where he lived was able to pull him out in time and so saved his life.
Does this mean their rescuers should be scolded in such cases for doing the wrong thing? This is a complicated question, but I think a plausible answer is that, no, they should be praised for trying to do the right thing.
An intention-based moralist might phrase this by saying "their hearts were in the right place," even though the actions sometimes went badly. The model-free approach Suffering reduction can be seen as an optimal control problem: Choosing an action policy to minimize suffering.
One direct and computationally feasible approach to optimal control is reinforcement learning RL. We have some control over RL rewards that we experience through our evaluation of an action: We feel bad for doing the wrong thing and good for doing the right thing.
If our aim is to shape the behavior of others, we can praise them for actions that reduce suffering and criticize them for actions that increase suffering. Mariko is part of a family that earns its living by fishing in a small boat off the coast.
Mariko feels sorry for the fish but knows that her family will continue to kill them no matter what she does. She volunteers to do the fish killing for the family, because otherwise the family members would just leave the fish to an awful death by asphyxiation.
Mariko practices ikejime in an effort to destroy the fish brains as quickly as possible. Should we blame Mariko for those failure cases?
The initial state is getting a fish to kill, and the possible actions are Ikejime or Asphyxiate. Ikejime leads to one of two possible states: Eventually all of these states return back to another state of getting a fish to kill, at which point the process repeats indefinitely.
Assume no gamma discount factor but some finite time horizon such that the expected values are finite. We can see this from the fact that Q is essentially just an exponential moving average with a smoothing factor equal to the learning rate.
Because the Q value 0. Of course, real humans may not use exactly Q-learning, but the general intuition remains the same: The model-based approach The RL approach works under the assumptions just mentioned: Accurate calibration of reward between individual and utilitarian evaluations and sufficient samples from which to learn.
This is not a case where you can try the actions multiple times and converge on the best response. Model-free RL has nothing to say here. Instead, we need to use our prior knowledge and models of the world to compute an optimal policy.
As a parallel to RL, this could be done using a dynamic programming approach for a Markov decision process. The goal is not to learn but to apply the given model to the situation at hand.
When the probabilities are known with certainty, then the model is all we need, and attempting to learn would only introduce noise. You can choose to toss a fair coin if you wish. If it is tossed and comes heads, 1 extra chicken endures a life of suffering on a factory farm.
If it comes tails, 2 chickens are prevented from a life of suffering on a factory farm. In this case, we should stick to the policy of flipping the coin. Even if when we do so, it comes up heads, this should not deter us from what we know is the right policy in the long run.
Of course, humans are naturally reinforcement-trained creatures.
Even with the coin-flipping case, after enough successive heads, the flipper should begin to doubt whether the coin is indeed fair though I assumed that away in the example. Thus, we should indeed keep learning from what we see happen.The words "moral" and "ethics" (and cognates) are often used interchangeably.
no right and wrong behaviour.Ý Although there are related claims that religion is necessary to motivate and guide people to behave in morally good way, most take the claim of the necessary connection between morality and religion to mean that right and wrong .
The words "moral" and "ethics" (and cognates) are often used interchangeably. However, it is useful to make the following distinction: Morality is the system through which we determine right and wrong conduct -- i.e., the guide to good or right conduct.. Ethics is the philosophical study of Morality..
What, then, is a moral theory? Toggle navigation Foundational Research Institute. Philosophy. Our Mission; "if you confront the universe with good intentions in your heart, it will reflect that and reward your intent.
Usually." consequentialists look like their moral evaluations are based on intentions. 1 For example. According to Kantian ethics, good intentions have the most value and are the necessary component for moral action.
For example, if a man finds an iPhone and thinks it is right to return it. For example, if a man finds an iPhone and thinks it is right to return it. In Kant’s terms, a good will is a will whose decisions are wholly determined by moral demands or, as he often refers to this, by the Moral Law.
Human beings inevitably feel this Law as a constraint on their natural desires, which is why such Laws, as applied to human beings, are imperatives and duties.
Good intentions are purposes or goals, something a person means to do that is right. According to Kantian ethics, good intentions have the most value and are the necessary component for moral action.