Reviewed learning material · Video analysis · EnglishRead the full overview
This video segment explains a college entrance exam probability problem inspired by the 2026 CCTV Spring Festival Gala robot performance, with Markov chains as the core concept. The instructor first emphasizes the importance of this topic in the college entrance exam, then carefully analyzes the problem conditions: robots of types A and B have different success/failure probabilities, and the robot type used in the next round depends on whether the current round succeeds (same type continues) or fails (type switches). When computing the mathematical expectation of the total score X over the first 3 rounds, the instructor stresses that one must first list all possible values of X(0,1,2,3), and demonstrates in detail how to compute P(X=0) using the state transition rules. For P(X=1), the instructor specifically corrects a common misconception: because the trials across rounds are not independent and identically distributed, one cannot directly apply the binomial coefficient C(3,1); instead, one must separately discuss and sum the three specific paths: "success-failure-failure", "failure-success-failure", and "failure-failure-success".
This segment explains a comprehensive probability problem about robot trials. Part (i) asks for the probability distribution of the total score over the first 3 rounds. The instructor enumerates all possible success–failure paths (e.g., A fails, B fails, A fails, etc.) and writes detailed probability multiplication expressions using the multiplication principle, emphasizing that such problems involve heavy computation and are prone to errors. Part (ii) shifts to finding the probability of using a specific robot model in round n. The instructor points out this is a typical Markov chain problem—where the current state depends only on the previous state—and guides viewers to use the law of total probability to establish a recurrence relation between Pn and Pn−1.
This segment explains a probability problem concerning robot model selection. First, based on the problem statement, the instructor writes down the total probability formula, yielding the first-order linear recurrence relation Pn=103Pn−1+21. Then, applying the method of undetermined coefficients, the instructor sets Pn+x=103(Pn−1+x) to solve for x=−75, successfully transforming the original sequence into a geometric sequence. Finally, using the initial condition P1=1, the closed-form formula for the new sequence is found and reverted to obtain Pn=75+72(103)n−1, completing the solution to part (a).
The video segment shows the teacher explaining a probability problem involving robot trials. The focus is on clarifying the concept of 'expected total score': using the law of total probability to compute the expected score per round, simplifying it into an expression in terms of Pn, and finally obtaining the structure of the sum Sn of the first n terms.
This segment illustrates the solution approach to part (ii) of a Markov chain probability problem. The teacher points out that to find the expected total score Sn over the first n trials, one may directly sum both sides of the recurrence formula for probabilities obtained in part (i). By decomposing Sn into a combination of a geometric series sum and a constant-term sum, the final expression is obtained.
Use the learning inspector for key ideas and moments, or open the reading tabs for the complete notes.
Generated from the video's visuals and explanation; not verbatim speech.
The video opens by directly identifying the core concept: Markov chains. The instructor notes that this topic has appeared twice in the national exam papers, highlighting its high frequency and importance, and urges students not to rely on luck but to master it thoroughly.
The instructor moves into problem analysis. He reads aloud and explains the experimental setup: a stability test involving robots of two types, A and B. The key rule is state transition: if the current type succeeds, the same type is used in the next round; if it fails, the other type is used instead. Specific probability parameters are given: type A has success probability 54 and failure probability 51; type B has both success and failure probability 21. Scoring rules: success earns 1 points, failure earns 0 points, and type A is fixed for round 1.
For part one, "find the mathematical expectation of the total score over the first 3 rounds of trials", Xthe teacher first emphasized the importance of writing solutions rigorously: one must explicitly list all possible values of the random variable E(X).XSince there are3each trial, the score is either0or1soXthe possible values of0,1,2,3areP(X=0)Next, the teacher derives, which is the probability that all three trials fail.
A detailed demonstration of computing P(X=0), illustrating the characteristic of Markov chains. Round 1 uses type A, with failure probability 51; due to failure, round 2 switches to type B, with failure probability 21; due to another failure, round 3 switches back to type A, with failure probability 51. By the multiplication principle, P(X=0)=51×21×51.
The analysis shifts to P(X=1), i.e., the case where exactly one success occurs among the three rounds. The instructor specifically points out a very common error: one must not directly use the binomial formula C31p(1−p)2. The reason is that the robot type used in each round—and thus the success probability p—depends on the outcome of the previous round, so the trials across rounds are neither independent nor identically distributed. The correct approach is to discuss three paths separately: "success-failure-failure", "failure-success-failure", and "failure-failure-success", computing for each path the corresponding type transitions and probabilities, then summing them up. The video ends here without showing the full calculation.
The video first addresses part one, which is to find the probability distribution of3the total score after n trials.XThe instructor carefully breaks down on the blackboard the specific paths forXdifferent values of n. For example, when calculatingP(X=1)n equals three, several cases are listed, such as 'correct-incorrect-incorrect' and 'incorrect-correct-incorrect', and according to the problem's rule—using the same model upon success and switching models upon failure—the corresponding probability products are written, such as54×51×21and so on.
Next, he analyzes the case of X=2, meaning exactly two successes among three rounds. Verbally, he lists combinations like 'success–success–failure' and 'failure–success–success', and attempts on the board to write the corresponding probability expressions, stressing that the success probability for each round (either 4/5 or 1/2) must be determined according to the specific model-switching state at that round. For X=3—three consecutive successes—the robot always remains on model A, so the probability is (54)3.
After completing the setup for part (i), the instructor pauses numerical computation and comments on the nature of such problems. He notes that although the logical reasoning involved is not exceptionally demanding ('no high cognitive load'), the problem requires extensive case analysis and fractional multiplication, resulting in heavy computation and a high risk of error, urging students to maintain patience.
Next comes part two, whose goal is to find the general formula fornthe probability that an A-model robot is used in the nth trial.PnThe instructor explicitly identifies this as a Markov chain model, whose key feature is that the state at step n depends only on the state at step n minus one, and not on any earlier history.nnn−1n minus one
The instructor begins deriving the recurrence relation. He reasons: if type-A robots are used in round n (an event with probability Pn), there are two possible origins: first, type-A robots were already used in round n−1 and succeeded (contributing probability Pn−1×54); second, type-B robots were used in round n−1 (with probability 1−Pn−1) and failed, thereby switching to type-A (contributing probability (1−Pn−1)×21). This yields the equation Pn=54Pn−1+21(1−Pn−1).
Let us analyze part (a) of this probability problem. The problem asks for the closed-form formula of the probability Pn that an A-model robot is used in round n.
According to the law of total probability, there are two ways for device A to be used on the nth trial: first, device A was used on the (n−1)th trial and is used again on the nth trial (with probability 54Pn−1); second, device B was used on the (n−1)th trial and device A is switched to on the nth trial (with probability 21(1−Pn−1)).
Adding these two parts and simplifying yields a first-order linear recurrence relation: Pn=103Pn−1+21, noting that n≥2.
To solve this recurrence, we apply the method of undetermined coefficients. Assume Pn+x=103(Pn−1+x); after expanding and comparing constant terms, we obtain −107x=21, solving which gives x=−75.
Substituting the value of x transforms the original recurrence into a geometric sequence form: Pn−75=103(Pn−1−75). This implies that the sequence {Pn−75} is a geometric sequence with common ratio 103.
Next, find the first term. Since device A is certainly used in the first round, P1=1. The first term of the new sequence is P1−75=1−75=72.
Applying the general term formula for geometric sequences yields Pn−75=72(103)n−1. Rearranging gives the final expression for Pn: Pn=75+72(103)n−1, where n∈N∗.
Thus, the first part of the problem is successfully solved. We have converted a complex probability problem into a familiar sequence problem and found its general term using the method of undetermined coefficients.
Hello everyone! Today, we continue exploring this major probability problem about robot experiments. The problem asks us to find an expression for the expected total score Sn over the first n rounds in terms of n. Many students feel confused upon seeing the phrase "expected total score" and do not know where to begin. In fact, we just need to return to the most fundamental concept: what does Pn represent? It represents the probability that the robot of type A is used in round n.
So, in round n, for the robot to earn a point, two conditions must both be satisfied: first, it must be selected for use; second, it must answer correctly. This requires us to apply the law of total probability. If the robot of type A is used in round n, the probability is Pn, and its probability of answering correctly is 54; if the robot of type B is used instead, the probability is 1−Pn, and its probability of answering correctly is 21.
Therefore, the expected score in round n is simply the weighted average of these two cases, namely Pn×54+(1−Pn)×21. Notice how this breaks down the complex "total score" into the "expected score per round". Once we compute the expected score for each round, the total expected score over the first n rounds is naturally just the sum of those individual expectations.
Next, we simplify this expression. Expanding the parentheses yields 54Pn+21−21Pn. Combining like terms, the coefficient of Pn becomes 54−21=108−105=103. Thus, the expected score in round n simplifies to 103Pn+21. This step is crucial, as it separates the probability term Pn from the constant term.
Finally, the problem asks for the sum of the first n rounds.SnBy the linearity of expectation, we can directly add up the expectations of each round.Sn=∑i=1n(103Pi+21)Factoring out the constant term 21 and summing it n times gives 2n, and the remaining part is 103(P1+P2+⋯+Pn)Thus, we have successfully transformed Sn into a summation problem involving the sequence {Pn}.
The video displays a probability problem involving a Markov chain. It is given that Pn is the probability that an A-type robot is used in trial n; part (i) has yielded the closed-form formula Pn=103Pn−1+21, which is further simplified into geometric series form Pn−75=103(Pn−1−75).
For part two, find the expected total score for the first n rounds of trials Sn, the teacher provided a summation approach. He pointed out that there is no need to re-derive a complicated probability model; instead, directly sum both sides of the recurrence relation obtained in part one.
The specific procedure uses the linearity of summation: summing the term 103Pn−1 on the right-hand side yields 103 times the series sum (involving a geometric series sum), while summing the constant term 21 for n times gives 2n directly.
The teacher notes that the algebraic simplification is quite tedious and recommends that students complete the calculation themselves after class to obtain the final expression of Sn in terms of n.
Knowledge cards
01
Application of Markov chains in college entrance exam probability problems
The video explains that when the state of the next trial (e.g., device model used, success probability) depends only on the outcome of the current trial, such problems belong to the Markov chain model. In the Gaokao, this commonly appears as the rule: 'continue using the same model upon success; switch models upon failure.' The key to solving such problems lies in clearly mapping out the state transition paths—never blindly apply the formula for independent repeated trials (binomial distribution).
02
Standard notation for the probability distribution table of a discrete random variable
Computing the mathematical expectation E(X) requires first determining the distribution table. As emphasized by the instructor in the video, the first step must be to list all possible values of the random variable X (for example, in this problem, the score after 3 rounds may be 0, 1, 2, or 3). Although drawing the full table is sometimes unnecessary, explicitly stating the range of possible values is a scoring point; omitting it leads to lost marks.
E(X)=∑xiP(X=xi)
03
Caution: Misusing the combination number C(n,k) under non-identical and non-independent distributions
This is the key pitfall stressed in this segment. When computing the probability of 'exactly k successes in n trials,' if the success probability differs across trials or if dependencies exist among trials (such as model switching in this problem), one must not use Cnkpk(1−p)n−k. Instead, one must enumerate all specific sequences satisfying the condition (e.g., 'success-failure-failure', 'failure-success-failure', etc.), compute each sequence's probability individually, and sum them.
04
Computing the probability of a specific path using the state transition rule
Taking P(X=0) (three consecutive failures) as an example, the video demonstrates how to deduce the probability at each step by applying the problem's rules: Round 1: A fails (51) → o triggers switching to o; Round 2: B fails (21) → o triggers switching to o; Round 3: A fails (51). Multiplying the probabilities of all steps yields the probability of this specific path. This method applies to all path-counting problems based on Markov chains.
P(X=0)=51×21×51
05
Calculation strategy for the probability distribution table of a discrete random variable
When handling the distribution of total scores across multi-stage experiments, the key is to enumerate all possible "success/failure" paths. The probability of each path equals the product of the conditional probabilities at each stage along that path. Note the state-transition rules specified in the problem (e.g., in this problem, failure changes the parameters for the next stage); one must not blindly apply the binomial distribution formula.
P(X=k)=paths with k successes∑∏P(stepi)
06
Memoryless property of Markov chains
Markov chains describe a stochastic process where the probability distribution of the system's state at a future time depends only on its current state, and not on any past states. In solving this problem, this means we need only usePn−1 to represent Pn, thereby transforming complex probability problems into sequence recurrence problems.
P(Xn+1∣Xn,Xn−1,...)=P(Xn+1∣Xn)
07
Establishing recurrence relations using the law of total probability
To compute the probability of being in a certain state at step n, one typically considers all possible states at step n−1. In this problem, the probability of using type-A robots at step n equals (the probability of using type-A robots at step n−1 and succeeding) plus (the probability of using type-B robots at step n−1 and failing). This is the standard modeling approach for such transition-probability problems.
Pn=P(An−1)P(An∣An−1)+P(Bn−1)P(An∣Bn−1)
08
Using the law of total probability to establish a recurrence relation
When handling probability problems involving multiple-state transitions, the law of total probability is often used to establish a recurrence relation between consecutive terms. In this problem, the state on the nth trial depends on the state on the (n−1)th trial; by case analysis—considering whether device A or device B was used in the previous round—we can write an equation relating Pn and Pn−1.
Pn=54Pn−1+21(1−Pn−1)
09
Constructing a geometric sequence via the method of undetermined coefficients
For a first-order linear recurrence sequence of the form an=pan−1+q (p=1) , we can solve for the constant an+x=p(an−1+x) by setting x. Expanding and comparing coefficients yields x=1−pq, thereby transforming the original sequence into a geometric sequence with common ratio p.
Pn+x=103(Pn−1+x)⟹x=−75
10
Applying the general term formula for geometric sequences
After constructing a geometric sequence, first find its first term and common ratio, then use formula bn=b1⋅qn−1 to write the general term, and finally revert back to the expression of the original sequence an. In this problem, special attention must be paid to using the initial condition P1=1.
Pn−75=72(103)n−1⟹Pn=75+72(103)n−1
11
Definition and computational strategy for expected total score
In this problem, "the expected total score over the first n rounds" does not refer to the average value of a single specific experimental outcome, but rather to the cumulative sum of the expected scores for each individual round. The core of the calculation lies in applying the law of total probability: first compute the expected score for round n, then sum over all rounds.
En=Pn⋅P(ACorrect)+(1−Pn)⋅P(BCorrect)
12
Algebraic simplification of the expected score for round n
After substituting the probabilities of using model A and model B in round n, along with their corresponding correct-answer probabilities, into the formula, algebraic simplification is required. The key step is to combine terms containing Pn, perform common-denominator subtraction on their coefficients, thereby obtaining a concise expression of the form k⋅Pn+b, which facilitates subsequent summation.
54Pn+21(1−Pn)=103Pn+21
13
Use the linearity of expectation to compute the sum of the first n terms.
Once the expected score for round i is found to be 103Pi+21, the total expected score Sn over the first n rounds is simply the sum of these n expressions. Using the commutative and associative laws of addition, the constant term 21 can be summed n times to yield 2n, and the terms involving Pi can be factored by extracting the common factor 103, reducing them to the sum of the first n terms of the sequence {Pn}.
Sn=i=1∑n(103Pi+21)=103i=1∑nPi+2n
14
Using a recurrence relation to compute the sum of the first n terms
When a sequence {an} satisfies a recurrence relation of the form an=k⋅an−1+c, and the sum of the first n terms, Sn, is required, one may directly sum both sides of the recurrence relation. Using the linearity of summation, ∑an=k∑an−1+∑c, one thus establishes an equation relating Sn to Sn−1 or to Sn itself, and then solves for Sn using the geometric series sum formula.
Sn=i=1∑nai
15
Summation method for expected total score
Sum both sides of the general term formula Pn=103Pn−1+21 over n. Using the linearity of summation, express Sn as 103Sn plus the sum of the constant term 21 taken n times, i.e., 2n.
Detailed learning notes
Explore conditions, steps and evidence. Supplementary explanations are labeled separately from content shown in the video.
Symbols · 17
A
Clear evidence
Shown in the video
Evidence
Formula
Observation
The problem statement reads: "The probability that model A robot trial succeeds is 4/5, and the probability it fails is 1/5."
Symbol
A
Meaning
Model A robot
Domain
discrete state
B
Clear evidence
Shown in the video
Evidence
Formula
Observation
The problem statement reads: "The probability that model B robot trial succeeds is 1/2, and the probability it fails is 1/2."
Symbol
B
Meaning
Model B robot
Domain
discrete state
X
Clear evidence
Shown in the video
Evidence
Formula
Observation
The problem statement reads: "Let X denote the total score over the first 3 trials."
Animation
Observation
The teacher writes on the whiteboard: "X can take values 0, 1, 2, 3."
Symbol
X
Meaning
total score over the first 3 trials
Domain
zero, one, two, three
P
Clear evidence
Shown in the video
Evidence
Formula
Observation
The teacher writes on the whiteboard: "P(X=0) = ..."
Symbol
P
Meaning
probability
Domain
zero to one
E(X)
Clear evidence
Shown in the video
Evidence
Formula
Observation
The problem statement reads "Find the mathematical expectation E(X) of X."
Symbol
E(X)
Meaning
The mathematical expectation of a random variable X
Domain
real number
X
Clear evidence
Shown in the video
Evidence
Formula
Observation
The problem statement defines "Let X be the total score from the first 3 trials."
Symbol
X
Meaning
total score from the first 3 trials
Domain
the set of possible values is {0, 1, 2, 3}
Pn
Clear evidence
Shown in the video
Evidence
Formula
Observation
The problem statement defines "Let Pn be the probability that an A-type robot is used in the nth trial."
Symbol
Pn
Meaning
the probability that an A-type robot is used in the nth trial
Domain
n is a positive integer
E(X)
Clear evidence
Shown in the video
Evidence
Formula
Observation
The problem statement asks to find "the mathematical expectation E(X) of X."
Symbol
E(X)
Meaning
the mathematical expectation of the random variable X
Domain
Not stated in the original text.
Pn
Clear evidence
Shown in the video
Evidence
Formula
Observation
Pn appears repeatedly in the board work and problem statement, representing the probability that an A-type robot is used in the nth trial.
Symbol
Pn
Meaning
the probability that an A-type robot is used in the nth trial
Domain
n is a positive integer
x
Clear evidence
Shown in the video
Evidence
Formula
Observation
In constructing a geometric sequence, the board work introduces an unknown x and writes Pn+x=103(Pn−1+x).
Symbol
x
Meaning
undetermined coefficient, used to convert a first-order linear recurrence into geometric sequence form
Domain
real number
Pn
Clear evidence
Shown in the video
Evidence
Formula
Observation
Pn appears multiple times in the board work and problem statement
Symbol
Pn
Meaning
the probability of using robot type A in the nth trial
Domain
n is greater than or equal to 1
Sn
Clear evidence
Shown in the video
Evidence
Formula
Observation
Sn appears in the problem statement
Symbol
Sn
Meaning
the expected total score over the first n trials
Domain
n is greater than or equal to 1
Knowledge points · 11
Markov chain
Clear evidence
Shown in the video
Evidence
Audio
Observation
The teacher mentions "Markov chains have appeared twice in the national exam" and "This robot probability problem tests Markov chains."
Definition
Explanation
The video points out that the core concept tested in this problem is a Markov chain, meaning that the state in the next round (which robot model is used) depends only on the current round's state and outcome (success or failure).
Formula
Conditions
The next round's state depends only on the current state
distribution table of a discrete random variable
Clear evidence
Shown in the video
Evidence
Audio
Observation
The teacher says "Pay attention to the possible values of X; omitting this sentence will result in lost points" and "It is best to list the distribution table."
Method
Explanation
Before computing the mathematical expectation of a discrete random variable, we must first determine all its possible values and compute the probability corresponding to each value, thereby constructing its distribution table.
Formula
Conditions
the random variable takes finitely many or countably infinitely many values
mathematical expectation
Clear evidence
Shown in the video
Evidence
Formula
Observation
The problem statement reads: "Find the mathematical expectation E(X) of X.",
Definition
Explanation
The mathematical expectation of a discrete random variable is the sum of the products of all its possible values and their corresponding probabilities.
Formula
E(X)=∑xiP(X=xi)
Conditions
the series converges absolutely
Prerequisites
distribution table of a discrete random variable
mathematical expectation of a discrete random variable
Clear evidence
Shown in the video
Evidence
Audio
Observation
The instructor says, "What does it mean when x equals 2? It means exactly 2 questions were answered correctly."
Formula
Explanation
The mathematical expectation is the sum of the products of all possible values of the random variable and their corresponding probabilities. In the video, the instructor lists the possible values of X—0, 1, 2, and 3—and computes their respective probabilities to prepare for calculating E(X).
Formula
E(X)=∑xiP(X=xi)
Conditions
X is a discrete random variable
Basic characteristics of a Markov chain
Clear evidence
Shown in the video
Evidence
Audio
Observation
The instructor explains, "What is a Markov chain? It means the state at step n is determined solely by the state at step n minus one."
Definition
Explanation
A Markov chain is a stochastic process whose defining feature is the "memoryless property": the state at step n depends only on the state at step n minus 1, and is independent of all earlier states.
Formula
Conditions
State transitions depend only on the previous state
first-order linear recurrence relation
Clear evidence
Shown in the video
Evidence
Formula
Observation
The board work writes Pn=54Pn−1+21(1−Pn−1) and simplifies it to Pn=103Pn−1+21(n≥2).
Audio
Observation
The teacher explains that, given a B-type robot was used at trial n-1, the conditional probability of using an A-type robot at trial n is 1/2.
Formula
Explanation
By the law of total probability, the probability of using an A-type robot at trial n equals the probability that an A-type robot was used at trial n-1 and is used again at trial n, plus the probability that a B-type robot was used at trial n-1 and an A-type robot is switched to at trial n.
Formula
Pn=54Pn−1+21(1−Pn−1)=103Pn−1+21(n≥2)
Conditions
n is greater than or equal to 2
Constructing a geometric sequence using the method of undetermined coefficients
Clear evidence
Shown in the video
Evidence
Formula
Observation
The board work demonstrates Pn+x=103(Pn−1+x), expands it, compares constant terms to obtain -107x=21, and solves for x=−75.
Audio
Observation
The teacher mentioned, 'If you don’t know the technique, use the method of undetermined coefficients: add x to both sides to construct a geometric sequence.'
Method
Explanation
For a recursive sequence of the form an=pan−1+q, we can solve for x by setting an+x=p(an−1+x), thereby transforming the original sequence into a geometric sequence.
Formula
Pn+x=103(Pn−1+x)
Conditions
p is not equal to 1
Prerequisites
first-order linear recurrence relation
General term formula for a geometric sequence
Clear evidence
Shown in the video
Evidence
Formula
Observation
The board work writes Pn−75=103(Pn−1−75) and finds the first term P1−75=72, ultimately deriving the general term formula.
Formula
Explanation
Using the general term formula for a geometric sequence, an=a1⋅qn−1, and combining the found first term and common ratio, obtain the general term of the new sequence, then recover the general term of the original sequence.
Formula
Pn=75+72(103)n−1(n∈N∗)
Conditions
n is a positive integer
Prerequisites
Constructing a geometric sequence using the method of undetermined coefficients
Method for calculating the expected total score
Clear evidence
Shown in the video
Evidence
Audio
Observation
The teacher explains the meaning of 'expected total score' and points out that we need to discuss cases separately: the probabilities of using model A and model B, and the probabilities of answering correctly.
Formula
Observation
The board displays: Pn⋅54+(1−Pn)⋅21
Method
Explanation
When calculating the expected score for round n, apply the law of total probability: multiply the probability of using model A by the probability of answering correctly with model A, then add the product of the probability of using model B and the probability of answering correctly with model B.
Formula
Pn⋅54+(1−Pn)⋅21
Conditions
The probability of using model A in round n is given as Pn
The probabilities of answering correctly with models A and B respectively are given
method for summing the expected total score
Clear evidence
Shown in the video
Evidence
Audio
Observation
The teacher says aloud: "This sum is its sum, then add another geometric series sum, and finally add n over two."
Formula
Observation
The whiteboard shows 103(P1+P2+⋯+Pn)+2n
Method
Explanation
Sum both sides of the general term formula Pn=103Pn−1+21 with respect to n; using the linearity of summation, rewrite Sn as 103Sn plus the sum of the constant term 21 multiplied by n, i.e., 2n.
Conditions
Given the recurrence relation for the sequence {Pn}
Geometric series sum
Clear evidence
Shown in the video
Evidence
Audio
Observation
The instructor says: "Next, a geometric series sum"
Definition
Explanation
During the derivation of Sn, since {Pn−75} is a geometric sequence with common ratio 103, the partial sum of this geometric sequence must be computed.
Conditions
The sequence is a geometric sequence
Derivations and proofs · 7
the process of computing P(X=0)
Clear evidence
Shown in the video
Evidence
Audio
Observation
The instructor explains: "X equals 0 means all three questions are incorrect... In the first trial, model A is used, and its failure probability is one-fifth... For the second robot, we switch to model B, whose failure probability remains one-half... For the next trial, we switch back to model A... and its failure probability remains one-fifth."
Animation
Observation
The whiteboard displays the expression P(X=0)=1/5×1/2×1/5
Proof
Steps
Expression
X=0⟹all three rounds fail
Explanation
A total score of 0 means all three trials fail.
Justification
The problem states that success earns 1 points and failure earns 0 points.
Shown in the video
Expression
round one: A fails
Explanation
In the first round, model A is used exclusively; the probability that model A fails is 1/5.
Justification
given condition in the problem
Shown in the video
Expression
round two: B fails
Explanation
Because round 1 fails, the rule requires switching to model B; the probability that model B fails is 1/2.
Justification
Given conditions and state transition rules.
Shown in the video
Expression
round three: A fails
Explanation
Because round 2 fails, the rule requires switching back to model A. The probability of A failing is 1/5.
Justification
Given conditions and state transition rules.
Shown in the video
Expression
P(X=0)=51×21×51
Explanation
Multiply the probabilities of failing in all three rounds to obtain the total probability.
Justification
Multiplication principle (chain multiplication of conditional probabilities).
Shown in the video
Conclusion
The probability that X equals zero is one fiftieth.
Case analysis approach for computing P(X=1).
Clear evidence
Shown in the video
Evidence
Audio
Observation
The teacher explains: "If X equals 1, it means you answered exactly one question correctly... namely success-failure-failure, success-failure-failure, success-failure-failure... You cannot have C31, because the trials are not independent and identically distributed."
Animation
Observation
On the whiteboard: P(X=1) = and partial expressions for the three cases: 4/5×1/2×1/2+1/5×1/2×4/5+1/5×1/2×1/5×4/5 (Note: The teacher did not finish writing the third case but verbally explained the logic.)
Uncertainties
The video cuts off at 150 seconds; the final calculation of P(X=1) remains incomplete.
Proof
Steps
Expression
X=1⟹exactly one success in the three rounds
Explanation
A total score of 1 means exactly one success and two failures among the first three trials.
Justification
The problem defines a success as worth 1 points and a failure as worth 0 points.
Shown in the video
Expression
case one: success, failure, failure
Explanation
In round 1, robot A succeeds (4/5); in round 2, robot A continues but fails (1/5); in round 3, robot B is used and fails (1/2). The probability is 4/5×1/5×1/2.
Justification
State transition rule: if successful, continue using the same model; if failed, switch to the other model.
Derived from the video
Expression
case two: failure, success, failure
Explanation
In round 1, robot A fails (1/5); in round 2, robot B is used and succeeds (1/2); in round 3, robot B continues but fails (1/2). The probability is 1/5×1/2×1/2.
Justification
State transition rule.
Derived from the video
Expression
case three: failure, failure, success
Explanation
In round 1, robot A fails (1/5); in round 2, robot B is used and fails (1/2); in round 3, robot A is switched back and succeeds (4/5). The probability is 1/5×1/2×4/5.
Justification
State transition rule.
Derived from the video
Expression
P(X=1)=∑probability of each case
Explanation
Add the probabilities of these three mutually exclusive cases.
Justification
Law of total probability / addition principle.
Shown in the video
Conclusion
We must consider three separate cases and sum their probabilities; we cannot use the combination notation C31 because the probabilities differ across rounds and depend on the current state.
Derivation of the recurrence relation for Pn
Clear evidence
Shown in the video
Evidence
Formula
Observation
The board displays Pn=54Pn−1+21(1−Pn−1)
Proof
Steps
Expression
Pn=54Pn−1+21(1−Pn−1)
Explanation
There are two cases in which model A is used at step n: first, model A was used at step n minus 1 and succeeded (with probability Pn−1×54); second, model B was used at step n minus 1 and failed (with probability (1−Pn−1) ×21).
Justification
The law of total probability and the state transition rules given in the problem
Shown in the video
Conclusion
Pn equals three-tenths times Pn−1, plus one-half.
Find the general term formula of the sequence {Pn}
Clear evidence
Shown in the video
Evidence
Formula
Observation
The complete board-work derivation process, from the recurrence relation to the final general term formula.
Proof
Steps
Expression
Pn=54Pn−1+21(1−Pn−1)
Explanation
Write down the law of total probability according to the problem statement.
Justification
Law of total probability
Shown in the video
Expression
Pn=103Pn−1+21
Explanation
Simplify the recurrence relation.
Justification
Algebraic manipulation
Shown in the video
Expression
Pn+x=103(Pn−1+x)
Explanation
Let the undetermined coefficient be x and construct a geometric sequence.
Justification
Method of undetermined coefficients
Shown in the video
Expression
−107x=21⟹x=−75
Explanation
Compare constant terms to solve for x.
Justification
Polynomial identity
Shown in the video
Expression
Pn−75=103(Pn−1−75)
Explanation
Substitute the value of x to obtain a new geometric recurrence relation.
Justification
Algebraic substitution
Shown in the video
Expression
P1=1⟹P1−75=72
Explanation
Compute the first term of the new sequence.
Justification
Problem statement: the first round must use model A.
Shown in the video
Expression
Pn−75=72(103)n−1
Explanation
Use the general term formula for a geometric sequence.
Justification
Property of geometric sequences
Shown in the video
Expression
Pn=75+72(103)n−1
Explanation
Rearrange terms to obtain the general term formula for {Pn}.
Justification
Algebraic manipulation
Shown in the video
Conclusion
The general term formula for the sequence {Pn} is Pn=75+72(103)n−1(n∈N∗).
Simplification of the expected score expression for round n
Clear evidence
Shown in the video
Evidence
Audio
Observation
The teacher says aloud: 'four fifths minus one half, eight tenths minus five tenths, three tenths times Pn'
Formula
Observation
The board displays the simplification steps
Proof
Steps
Expression
Pn⋅54+(1−Pn)⋅21
Explanation
Write down the original expression for the expected score in round n
Justification
Application of the law of total probability
Shown in the video
Expression
54Pn+21−21Pn
Explanation
Expand the parentheses
Justification
Distributive property of multiplication
Shown in the video
Expression
(54−21)Pn+21
Explanation
Combining like terms
Justification
Algebraic operation rules
Shown in the video
Expression
103Pn+21
Explanation
Computing coefficients to obtain the final simplified result
Justification
Fraction subtraction
Shown in the video
Conclusion
The expected score for round n is 103Pn+21
Summation for the total expected score Sn over the first n rounds
Clear evidence
Shown in the video
Evidence
Audio
Observation
The teacher says aloud, "For the sum of the first n terms, we must compute its sum—that is, three-tenths times P1 plus P2 plus Pn, and then plus n over two."
Formula
Observation
Writing the summation expression on the board
Proof
Steps
Expression
Sn=i=1∑n(103Pi+21)
Explanation
Adding the expected scores from the first n rounds
Justification
Linearity of expectation
Derived from the video
Expression
Sn=103(P1+P2+⋯+Pn)+2n
Explanation
Separating constant terms from terms containing Pi
Justification
Commutative and associative laws of addition
Shown in the video
Conclusion
Sn equals three-tenths times the sum of Pi from i equals 1 to n, plus one-half of n
Derivation of the expression for Sn
Clear evidence
Shown in the video
Evidence
Audio
Observation
The instructor verbally explains the summation steps
Formula
Observation
The expression after summation appears on the right side of the whiteboard
Proof
Steps
Expression
Pn=103Pn−1+21
Explanation
Write the probability recurrence formula for the nth term
Justification
Content already present on the video whiteboard
Shown in the video
Expression
i=1∑nPi=103i=1∑n−1Pi+i=1∑n21
Explanation
Sum both sides of the equation from 1 to n
Justification
linearity property of summation
Derived from the video
Expression
Sn=103(Sn−Pn)+2n
Explanation
replace the summation symbol with the form Sn
Justification
definition of Sn
Derived from the video
Conclusion
relate Sn to Pn via summation, thereby solving for the expression of Sn in terms of n
Worked examples · 3
A probability problem from the 2026 CCTV Spring Festival Gala involving robots.
Clear evidence
Shown in the video
Evidence
Formula
Observation
The screen displays the complete problem statement, including background, rules, probabilities, and the two questions.
Problem
To test the motion stability of two robot models, A and B, a laboratory designs the following experiment: each trial is independent and consists of performing one action. If a robot of a given model succeeds in the trial, the next trial uses the same model; if it fails, the next trial switches to the other model. The success probability for model A is 4/5, and its failure probability is 1/5; the success probability for model B is 1/2, and its failure probability is 1/2. A successful trial scores 1 points, and a failed trial scores 0 points. Robot A is used in round 1. (1) Let X be the total score over the first 3 trials; find the mathematical expectation E(X). (2) Let Pn denote the probability that model A is used in trial n. ① Find the closed-form expression for the sequence {Pn}; ② Let Sn be the expected total score over the first n trials; find an expression for Sn in terms of n.
Given
The probability of success for A is 4/5, and the probability of failure is 1/5.
The probability of success for B is 1/2, and the probability of failure is 1/2.
Each success earns 1 points, and each failure earns 0 points.
Model A is used in round 1.
State transition rule: continue using the same model after success; switch models after failure.
Goal
Find the mathematical expectation E(X) of the total score X over the first 3 rounds. (The video mainly explains the reasoning for part one.)
Steps
Expression
X∈{0,1,2,3}
Explanation
Determine all possible values of the random variable X.
Justification
There are 3 rounds of trials, and each round yields either 0 or 1 points.
Shown in the video
Expression
P(X=0)=51×21×51
Explanation
Calculate the probability that all three rounds result in failure.
Justification
Derive the model used and the failure probability for each round based on the state transition rule.
Calculate the probability of exactly one success, which requires discussing three separate cases.
Justification
The probabilities for each round differ and depend on previous states, so the binomial distribution formula cannot be applied.
Shown in the video
Answer
The video does not provide a final numerical answer, only showing the expression for computing P(X=0) and the case analysis framework for P(X=1).
Verification
By enumerating all possible state transition paths (such as A->A->A, A->B->B, etc.), we verify whether the probability calculation is complete.
Probability calculation for the robot experiment score
Clear evidence
Shown in the video
Evidence
Formula
Observation
The board shows detailed calculations for P(X=0), P(X=1), P(X=2), and P(X=3)
Problem
Given that model A has success probability 4/5 and model B has success probability 1/2; if a trial fails, switch to the other model; if it succeeds, continue with the same model. Start with model A in round one. Find the probability distribution of the total score X over the first 3 rounds.
Given
P(A succeeds) = 4/5, P(A fails) = 1/5
P(B succeeds) = 1/2, P(B fails) = 1/2
Initial state is A
Goal
Compute P(X=k) for k=0, 1, 2, 3
Steps
Expression
P(X=0)=51×21×21
Explanation
Three consecutive failures: A fails (1/5) → switch to B→B fails (1/2) → switch to A→A fails (1/5). Note: In the video’s board work, the final step is written as 1/5, but the spoken logic implies switching back to A and then A failing, which should be 1/5×1/2×1/5. The board work shows 1/5×1/2×1/2, possibly a handwriting error or a specific interpretation; here we record the board work faithfully.
Justification
Multiplication principle
Shown in the video
Expression
P(X=3)=54×54×54
Explanation
Three consecutive successes: always use A and all succeed.
Justification
Multiplication principle
Shown in the video
Answer
The video does not provide final numerical results, stopping at the stage of setting up expressions.
Verification
The original text does not display verification steps.
Robot model selection probability problem
Clear evidence
Shown in the video
Evidence
Formula
Observation
The full problem context appears at the top of the screen, involving state transition probabilities for robot models A and B.
Problem
Let Pn denote the probability of using model A in round n; find the general term formula for {Pn}.
Given
Model A success probability 4/5, failure probability 1/5
Model B success probability 1/2, failure probability 1/2
If successful, use the same model in the next round; if failed, switch to the other model in the next round
Use model A in round 1
Goal
Find the expression for Pn in terms of n
Steps
Expression
Pn=54Pn−1+21(1−Pn−1)
Explanation
Establish a recurrence relation using the law of total probability.
Justification
The state transition rule given in the problem
Shown in the video
Expression
Pn=103Pn−1+21
Explanation
Simplify the recurrence relation.
Justification
Algebraic manipulation
Shown in the video
Expression
Pn−75=103(Pn−1−75)
Explanation
Construct a geometric sequence using the method of undetermined coefficients.
Justification
Method of undetermined coefficients
Shown in the video
Expression
Pn=75+72(103)n−1
Explanation
Combine the initial condition P1=1 to find the general term formula.
Justification
General term formula for a geometric sequence
Shown in the video
Answer
Pₙ equals five sevenths plus two sevenths times three tenths raised to the power of n minus one, where n is a positive integer
Verification
When n=1, P1=5/7+2/7=1, which satisfies the problem condition.
Visual events · 5
Handwritten solution process on whiteboard
Clear evidence
Shown in the video
Evidence
Animation
Observation
The instructor handwrites the solution process on the whiteboard area on the right side of the screen, including "Possible values of X: 0, 1, 2, 3", "P(X=0)=1/5×1/2×1/5", and part of the expression "P(X=1) = ...".
Objects
Whiteboard
Handwritten text
Mathematical formula
Changes
Instructor writes formulas while explaining
Demonstrates the logic of case analysis
Invariants
The problem statement remains displayed on the left side of the screen throughout
Interpretation
Intuitively illustrates the steps for finding the probability distribution of a discrete random variable, especially how to compute the probability of a specific event using the state transition property of a Markov chain.
Whiteboard writing process
Clear evidence
Shown in the video
Evidence
Animation
Observation
The instructor writes probability calculation formulas and recurrence relations in real time on an electronic whiteboard.
Objects
Problem statement
Handwritten formula
Changes
Step-by-step expansion of the formula
Key symbols highlighted with red circles
Invariants
Background text of the problem remains unchanged
Interpretation
Demonstrating the progression of problem-solving reasoning through dynamic writing
Red-ink annotation of probability terms
Clear evidence
Shown in the video
Evidence
Animation
Observation
The teacher circles Pn−1 and (1−Pn−1) in red ink and uses arrows to indicate that they represent the probabilities of using A and B, respectively, on the (n-1)th trial.
Objects
P subscript n minus one
one minus P sub n minus one
Changes
add a red circle and an arrow
Invariants
the mathematical meaning of the formula itself remains unchanged
Interpretation
emphasize the physical meaning of each term in the law of total probability, namely the probability of using a different model in the previous round.
Whiteboard derivation demonstration
Clear evidence
Shown in the video
Evidence
Diagram
Observation
The right side of the screen displays the complete problem statement and handwritten derivation process; the left side shows a half-body portrait of the teacher
Objects
Problem statement
Handwritten formula
Teacher's portrait
Changes
As the explanation progresses, new derivation formulas and circled annotations are added step by step to the whiteboard
Invariants
The problem statement remains visible at all times
The teacher's portrait remains fixed in position
Interpretation
Real-time writing demonstrates the logical chain of mathematical derivation, helping viewers understand the process from probability to expectation to summation.
whiteboard derivation process
Clear evidence
Shown in the video
Evidence
Animation
Observation
the teacher circles the general term formula of Pn in red on the whiteboard and writes related summation expressions beside it
Objects
whiteboard
red handwriting
problem statement
Changes
key formula circled in red
add summation expressions
Invariants
problem statement remains unchanged
The left-side character image remains unchanged.
Interpretation
A visual demonstration of the thought process from the general term formula to the summation expression.
Misconceptions · 4
Misusing binomial distribution combinations
Clear evidence
Shown in the video
Evidence
Audio
Observation
The instructor emphasizes, "You cannot have C31 because the trials are not independent and identically distributed."
Misconception
consider3success in the trial1the probability of success in the trial can be directly calculated using C3 raised to the power of 1 times p times (1 minus p) raised to the power of 2.
Clarification
Since the robot model used in each trial—that is, the success probability—depends on the outcome of the previous trial, the trials are neither independent nor identically distributed; therefore, one must consider specific success/failure sequences case by case.
Ignoring computational workload leads to mental breakdown
Clear evidence
Shown in the video
Evidence
Audio
Observation
The instructor jokes, 'After finishing the problem, you just want to curse... The computational workload of recent college entrance exam problems is still quite heavy.'
Misconception
Underestimating the computational process because the problem logic seems simple
Clarification
Although the Markov chain logic is clear, computing the specific probability distribution involves numerous fraction multiplications, requiring patience and careful calculation
the range of values for n in the recurrence relation
Clear evidence
Shown in the video
Evidence
Audio
Observation
The teacher specifically emphasizes, "You may write n greater than or equal to 2."
Formula
Observation
The board writing annotates the recurrence relation with (n≥2).
Misconception
ignore the range of values for n in the recurrence relation and mistakenly assume it holds for all positive integers.
Clarification
Since the recurrence relation depends on the previous term Pn−1, it must satisfy n≥2.
Misconception about the concept of 'expected total score'
Clear evidence
Shown in the video
Evidence
Audio
Observation
The teacher repeatedly asks, "What does total score mean?", "Think about what Pn represents."
Misconception
Students may mistakenly think that 'total score' is simply the sum of individual scores, overlooking the underlying probability-weighted process.
Clarification
The 'expected total score' refers to the cumulative sum of the expected scores per round, where each round's expected score must be computed by weighting according to the probabilities of using different model robots in that round.
Concept relations · 6
Markov chain → distribution table of a discrete random variable
Clear evidence
Shown in the video
Evidence
Audio
Observation
The teacher points out that the problem tests Markov chains and applies state transition rules in solving it.
Application
Explanation
The state transition property of Markov chains is applied to compute the probabilities of the discrete random variable (total score) taking each possible value.
Basic characteristics of a Markov chain → Derivation of the recurrence relation for Pn
Clear evidence
Derived from the video
Evidence
Audio
Observation
The instructor identifies this problem as a Markov chain and lists the relationship between Pn and Pn−1
Application
Explanation
Use the Markov property (memorylessness) to establish a recurrence relation for the sequence.
first-order linear recurrence relation → Robot model selection probability problem
Clear evidence
Supplementary explanation
Evidence
Formula
Observation
The problem context is a typical Markov chain state transition, and the board writing transforms it into a first-order linear recurrence sequence.
Application
Explanation
A first-order linear recurrence relation is the core mathematical tool for solving such Markov chain probability problems.
Constructing a geometric sequence using the method of undetermined coefficients → General term formula for a geometric sequence
Clear evidence
Shown in the video
Evidence
Audio
Observation
The teacher says, "Here we construct a geometric sequence."
Application
Explanation
The method of undetermined coefficients is the key technique for transforming a general first-order linear recurrence sequence into a geometric sequence.
Method for calculating the expected total score → Simplification of the expected score expression for round n
Clear evidence
Shown in the video
Evidence
Audio
Observation
The teacher explains how to compute the expected score using Pn.
Formula
Observation
The board work shows the transformation from probability Pn to the expected-value expression.
Application
Explanation
The method for computing the expected total score (based on the law of total probability) is applied to derive the explicit expected-value expression for round n.
method for summing the expected total score → Geometric series sum
Clear evidence
Shown in the video
Evidence
Audio
Observation
The teacher mentions 'geometric series summation'.
Application
Explanation
In the method for computing the expected total score Sn, knowledge of geometric series summation is required.
Find an answer · 7
What is the frequency and question type of Markov chains in the college entrance examination mathematics test?
Clear evidence
Shown in the video
Evidence
Audio
Observation
The teacher mentions, "Markov chains have appeared twice in the national college entrance examination papers."
Knowledge points
Markov chain
Why can the binomial distribution combination formula not be used in probability problems involving state-dependent outcomes?
Clear evidence
Shown in the video
Evidence
Audio
Observation
The teacher explains why C31 cannot be used.
Knowledge points
Misusing binomial distribution combinations
Case analysis approach for computing P(X=1).
How to derive a recurrence formula for the probability of using a certain model in round n, based on the properties of a Markov chain?
Clear evidence
Shown in the video
Evidence
Audio
Observation
The instructor explains how to find the closed-form formula for Pn.
Knowledge points
Basic characteristics of a Markov chain
Derivation of the recurrence relation for Pn
How can we use the method of undetermined coefficients to transform a first-order linear recurrence sequence into a geometric sequence and find its closed-form formula?
Clear evidence
Supplementary explanation
Evidence
Formula
Observation
This entire segment derives the closed-form formula for Pn.
Knowledge points
first-order linear recurrence relation
Constructing a geometric sequence using the method of undetermined coefficients
General term formula for a geometric sequence
How should we interpret the phrase 'expected total score for the first n rounds' in the problem?
Clear evidence
Shown in the video
Evidence
Audio
Observation
The teacher asks, 'What does "total score" mean?'
Knowledge points
Method for calculating the expected total score
Misconception about the concept of 'expected total score'
How do we simplify the expected-score expression for round n and compute the sum of the first n terms?
Clear evidence
Shown in the video
Evidence
Formula
Observation
The board work shows the simplification and summation process.
Knowledge points
Simplification of the expected score expression for round n
Summation for the total expected score Sn over the first n rounds
How to compute the sum of the first n terms of a sequence using a recurrence formula?
Clear evidence
Shown in the video
Evidence
Audio
Observation
The teacher explains how to sum a recurrence relation.
Knowledge points
method for summing the expected total score
Derivation of the expression for Sn
Coverage and review notes
Covered · Introducing the Markov chain concept and the problem context
Covered · Reading the problem conditions to determine the possible values of the random variable X
Covered · The detailed process for computing P(X=0)
Covered · Analysis of the case-by-case reasoning and common pitfalls for P(X=1)
Covered · Part (i): Compute the probability distribution of the score X over the first 3 rounds; the instructor lists detailed multiplication expressions for probabilities of all cases.
Covered · Part (ii): Introduce the concept of a Markov chain and derive the recurrence formula for Pn.
Covered · Establish and simplify the total probability recurrence relation
Covered · Construct a geometric sequence using the method of undetermined coefficients
Covered · Solve for the closed-form formula of the geometric sequence and revert it to an expression for Pn
Covered · Part (a) concludes; we now proceed to part (b)
Covered · This segment fully covers the teacher's explanation of the meaning of 'expected total score', the derivation of the expected-value expression for round n, and the computation of the sum of the first n terms.
Covered · This segment fully covers the teacher's explanation of how to sum using a recurrence formula.
