Which person leads this area is very important.
That thinking is my origin, however, lots of people achieve nural network models.
// One of Cyber reports. // I've written them about A.I. // We depend on switch-case and if-else when we make computer judge something. // Same rate if you use "switch". switch ( a ) { case b: break; case c: break; case d: break; case e: break; case f: break; case g: break; case h: break; } // We can hit rate and improve params if we use "if". // Hitting patterns if ( a >= b && a < c ) { } else if ( a >= c && a < d ) { } else if ( a >= e && a < f ) { } else if ( a >= f && a < g ) { } else if ( a >= g && a < h ) { // go to judge-014. } // Once, create the above parameters. // 001, a, b:011, c:012, d:013, e:017, f:016, g:015, h:014 002, a, b:021, c:022, d:023, e:027, f:026, g:025, h:024 003, a, b:031, c:032, d:033, e:037, f:036, g:035, h:034 004, a, b:041, c:042, d:043, e:047, f:046, g:045, h:044 // Hitting pattern in array. int judge_num = 11; for ( int i=0; i<N; i++ ) { float[] f_ary = judge( judge_num ); int[] i_ary = judge( judge_num ); int num = sizeof (f_ary); for ( int j=1; j= f_ary[ j ] && f_ary[0] < f_ary[ j + 1 ] ) judge_num = i_ary[ j ]; break; } } // alphabet is float and improve it. // in nural network, once it pathes the root and increase // like the below. e += 0.0013943;
