Heuristics Vs Meta-heuristics

Heuristics and Metaheuristics are two terms, which are widely used in computer science. The definitions of the terms are unclear and the distinction between them vague. So in the post, I try to present my understanding of the same. The material provided below is a collection from different sources of the Internet.

Heuristics – For a given problem, when finding a solution through exhaustive search is impractical, we use a heuristic to find a solution. Heuristics can be “good guess” from the solution space. Heuristics are speedy and find an approximate solution. Heuristic solutions are found by trading optimality, completeness, accuracy and/or processing speed. Heuristics are often problem-dependent, that is, you define and heuristic for a given problem. A heuristic exploits problem-dependent information to find a ‘good enough’ solution to an specific problem

Metaheuristics – Metaheuristics make few assumptions about the optimization problem being solved, and so they usable for a variety of problems. Compared to simpler heuristics, metaheuristics are more abstract procedures that use low-level heuristics; thus, metaheuristics use concrete heuristics (or algorithms). Metaheuristics do not guarantee that a globally optimal solution can be found on some class of problems. Metaheuristics are problem-independent techniques that can be applied to a broad range of problems. A metaheuristic knows nothing about the problem it will be applied, it can treat functions as black boxes. Metaheuristics are, like design patterns, general algorithmic ideas that can be applied to a broad range of problems. While there is no commonly accepted definition for the term metaheuristic, there are properties that characterize most metaheuristics:

• Metaheuristics are strategies that guide the search process.
• The goal is to efficiently explore the search space in order to find near–optimal solutions.
• Techniques which constitute metaheuristic algorithms range from simple local search procedures to complex learning processes.
• Metaheuristic algorithms are approximate and usually non-deterministic.
• Metaheuristics are not problem-specific.

Read Full Post »

Ek kavi hoon, ek kavitha ki zaroorat hai

Presenting my latest poem, one of the rare poems, I have penned in Hindi. I have translated into English to the best possible extent. This is a poet’s imagination about the kind of beloved he would like to have in his life.
Wo mashuka apni ho
(May such be my beloved)

Jiske seerat aur surat mein sadhgi ho
Jiski chehre ki noor se meri jeevan ujwal ho
(Whose inner and outer beauty be pure and serene
May my life light up with the radiance of her face)

Jiske alfazon mein ruhaniyat jhalke
Jiske badan se bhi itr mehke
(Let her words sparkle divinity and spirituality
May holy fragrance ooze out of her personality)

Jiske aankhon mein duniya ke gham doobjaaye
Jiske anchal mein sansar ka saransh miljaaye
(Let the sorrowness of the whole word dissolve in her eyes
May I get the gist of the world in her lap)

Jiski khayalat se kaynat ko rah mile
Aisi mashuka humme mile
(May the universe find direction from her intellect,
Let such a person be my beloved)

Ek kavi hoon, ek kavitha ki zaroorat hai.
(I am poet, and all I need is a poetry.)

Read Full Post »