It might be helpful if you showed us what sort of data you are looking at since copy number variations and gene expression can each be detected in similar but distinct ways.
Generally speaking, both gene expression and CNVs are detected by counting the number of reads that map to a given genomic region. The difference is that CNVs are detected using DNA sequencing while gene expression is detected using RNA sequencing. This is an important distinction because CNVs are a more stable genomic feature than gene expression. Reviewing the Central Dogma might help if this is confusing.
In the context of your problem, the main thing to understand is that CNVs can exert an influence on gene expression but gene expression cannot exert an influence on CNVs. Having fewer copies of a gene means that the process of transcribing RNA to express it cannot proceed as quickly as normal. So in an organism with gene X showing copy number loss, it makes sense to expect diminished expression of gene X than in organisms whose genome includes more copies of that gene X. If you're asking for an undergraduate class, chances are this is all you're expected to understand about "copy number regulation": the maximum transcription rate of a gene is dependent on how many copies of that gene are available for transcription.
Gene regulation becomes important for understanding how an organism with fewer than normal copies of gene X in its DNA might have the same amount of gene X RNA expressed as an organism with a normal number of copies of gene X. Suppose that some molecule, M, is necessary for expression of X but degrades after a short time. The cell will send M to the nucleus whenever it needs X to be expressed and the number of X transcripts produced per molecule of M will be dependent on the number of X copies available for transcription. If M is abundant and can be continually sent to the nucleus, then the cell with fewer copies of X can theoretically produce as much X as the cell with a normal number of X copies although it'll take a bit longer. If instead M is a finite resource within the cell, then the amount of X expressed in cells with fewer copies of the gene should be consistently lower than cells with more copies.
Of course if M happens to be produced by a gene exhibiting copy number gain/loss then that could also play an important role in how much X RNA transcripts you ultimately detect (even in the event that X actually has a neutral copy number). Hope this helps.