move from single candidate td-leaf with gating to tournament of a population

The goal in this change was to overcome the rock-paper-scissors phenomenon that was happening in capability space by having on a single candidate learn to beat the most recent version of itself. In some sense this lead to rotating around in capability space instead of chasing monotonic improvements.

I hope that flipping to a system in which many (N=5) candidates play each other and learn from their own set of diverse games in this pool, combined with --replay-iters, will lead to a given network having to have a much more comprehensive understanding of the space of tactics. I haven't yet had time to train this thorougly, will run those experiments on my desktop computer.

1 files changed, 6 insertions, 6 deletions

diff --git a/include/nn.c b/include/nn.c
index 909b870..9b241a1 100644
--- a/include/nn.c
+++ b/include/nn.c
@@ -139,17 +139,17 @@ float eval_net(const float *board, const unsigned int n_inp, const float *w,
 float five_nn_fn(tak_state_p state) {
   static float cur_board[FIVE_INP_NUM] NN_ALIGNED;
 
-  nn_prepare_input(cur_board, state, five_embedding);
-  return eval_net(cur_board, FIVE_INP_NUM, &five_dense_weights[0][0],
-                  five_dense_biases, five_logit_weights, five_logit_bias,
+  nn_prepare_input(cur_board, state, FIVE_embedding);
+  return eval_net(cur_board, FIVE_INP_NUM, &FIVE_dense_weights[0][0],
+                  FIVE_dense_biases, FIVE_logit_weights, FIVE_logit_bias,
                   FIVE_DENSE_NUM, state->ply & 1 ? 1.0f : -1.0f);
 }
 
 float six_nn_fn(tak_state_p state) {
   static float cur_board[SIX_INP_NUM] NN_ALIGNED;
 
-  nn_prepare_input(cur_board, state, six_embedding);
-  return eval_net(cur_board, SIX_INP_NUM, &six_dense_weights[0][0],
-                  six_dense_biases, six_logit_weights, six_logit_bias,
+  nn_prepare_input(cur_board, state, SIX_embedding);
+  return eval_net(cur_board, SIX_INP_NUM, &SIX_dense_weights[0][0],
+                  SIX_dense_biases, SIX_logit_weights, SIX_logit_bias,
                   SIX_DENSE_NUM, state->ply & 1 ? 1.0f : -1.0f);
 }