One way to improve a model is to change one of these settings in the Run settings section:
Batch size, the batch size is the number of training examples in one iteration.
Number of epochs, an epoch is when all training data have run through the model once. You might get a better result if you let the experiment run for a longer time.
Optimizer, this is how the system optimizes the loss with respect to the weights of the network.
Learning rate, a high learning rate means faster adaptation but at the cost of higher variance in the training. A too low learning rate means that you might get stuck at a local minimum.
You can also change the parameters in specific blocks to experiment with the model for better performance, for example, change Activation or number of Nodes in a Dense block.