What Deep Learning can not do?

Deep learning is great! But it also has limitations. 

1. It need accurate data (or at least not so dirty data);
2. It usually needs large amount of data or many many tries;

e.g., by contrast, humans “learn from very few examples, can do very long-term planning, and are capable of forming abstract models of a situation and manipulate these models to achieve extreme generalization.”

Even simple human behaviors are laborious to teach to a deep learning algorithm. Let’s examine the task of not being hit by a car as you walk down the road. If you go the supervised learning route, you’d need huge data sets of car situations with clearly labeled actions to take, such as “stop” or “move”. Then you’d need to train a neural network to learn the mapping between the situation and the appropriate action.

If you go the reinforcement learning route, where you give an algorithm a goal and let it independently determine the ideal actions to take, the computer would need to die thousands of times before learning to avoid cars in different situations.

“You cannot achieve general intelligence simply by scaling up today’s deep learning techniques,” warns Francois Chollet.

3. As for success of AlphaGo, it's not purely based on deep learning. It is built with 
deep supervised learning + deep reinforcement learning + Monte Carlo Tree Search
each part is essential to make it successful.  

4. Reasoning and logical thinking, see following example. (all machine learning algorithms can't do it) 
quoted from LINK