Tensorflow最小化导致损失最大化？

Question

此代码如何：

def policy_gradient():
    with tf.variable_scope("policy"):
        params = tf.get_variable("policy_parameters",[4,2])
        state = tf.placeholder("float",[None,4])
        actions = tf.placeholder("float",[None,2])
        advantages = tf.placeholder("float",[None,1])
        linear = tf.matmul(state,params)
        probabilities = tf.nn.softmax(linear)
        good_probabilities = tf.reduce_sum(tf.mul(probabilities, actions),reduction_indices=[1])
        eligibility = tf.log(good_probabilities) * advantages
        loss = -tf.reduce_sum(eligibility)
        tf.summary.scalar("loss", loss)
        optimizer = tf.train.AdamOptimizer(0.01).minimize(loss)

        log_ph = tf.placeholder("float", [])
        tf.summary.scalar("totalreward log", log_ph)
        return probabilities, state, actions, advantages, optimizer, log_ph

产生这个情节？

奖励正在正确地增加，并且实现是openai的官方解决方案，所以我很确定该算法是正确的...

来源：https://github.com/kvfrans/openai-cartpole/blob/master/cartpole-policygradient.py