State graphs and transitions in Agentic AI - Practice Problems & Coding Challenges

Challenge - 5 Problems

🧠 Conceptual

intermediate

Understanding State Transitions in Agentic AI

In an agentic AI system, a state graph represents possible states and transitions between them. Which statement best describes a deterministic transition?

A Each state has exactly one possible next state for a given action.

B A state can transition to multiple next states randomly for the same action.

C Transitions depend on external random noise, making outcomes unpredictable.

D States loop back to themselves without any change.

💡 Hint

Think about whether the next state is fixed or can vary for the same action.

Deterministic transitions mean that for a given state and action, the next state is always the same, with no randomness involved.

❓ Predict Output

intermediate

Output of State Transition Function

Given the following Python function representing state transitions, what is the output of next_state('S1', 'a')?

Agentic AI

def next_state(current_state, action):
    transitions = {
        'S1': {'a': 'S2', 'b': 'S3'},
        'S2': {'a': 'S2', 'b': 'S1'},
        'S3': {'a': 'S1', 'b': 'S3'}
    }
    return transitions.get(current_state, {}).get(action, 'Invalid')

print(next_state('S1', 'a'))

A 'S3'

B 'Invalid'

C 'S1'

D 'S2'

💡 Hint

Check the dictionary for state 'S1' and action 'a'.

The dictionary maps 'S1' with action 'a' to 'S2', so the function returns 'S2'.

❓ Hyperparameter

advanced

Choosing Transition Probability in Stochastic State Graphs

In a stochastic state graph, transitions have probabilities. If you want the agent to explore more states randomly, which transition probability setting is best?

A Assign high probability to one transition and near zero to others.

B Assign equal probabilities to all possible transitions.

C Assign zero probability to all transitions except one.

D Assign probabilities based on the shortest path only.

💡 Hint

Think about how equal chances affect exploration.

Equal probabilities encourage the agent to explore all possible next states rather than favoring one.

❓ Metrics

advanced

Evaluating State Transition Model Accuracy

You have a model predicting next states in a state graph. Which metric best measures how often the model predicts the correct next state?

A Accuracy

B Precision

C Mean Squared Error (MSE)

D Recall

💡 Hint

Consider a metric that counts correct predictions over total predictions.

Accuracy measures the fraction of correct next state predictions out of all predictions.

🔧 Debug

expert

Debugging State Transition Code

What error does the following code raise when calling get_next_state('S1', 'c')?

def get_next_state(state, action):
    transitions = {
        'S1': {'a': 'S2', 'b': 'S3'},
        'S2': {'a': 'S2', 'b': 'S1'},
        'S3': {'a': 'S1', 'b': 'S3'}
    }
    return transitions[state][action]

print(get_next_state('S1', 'c'))

A TypeError

B IndexError

C KeyError

D No error, returns None

💡 Hint

Check what happens if the action key does not exist in the dictionary.

Accessing a missing key in a dictionary raises a KeyError in Python.