MathOverflow is a question and answer site for professional mathematicians. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

Let $\mathcal A, \mathcal B$ be dg-categories over a field $k$. I denote by $\mathcal{RHom}(\mathcal A,\mathcal B)$ the dg-category (defined up to quasi-equivalence) which gives the internal hom in $\mathrm{Hqe}$ (see, for example, Keller's survey, Theorem 4.5). There is a natural $H^0$ functor: \begin{equation} H^0(-) :H^0(\mathcal{RHom}(\mathcal A,\mathcal B)) \to \mathcal{Hom}(H^0(\mathcal A),H^0(\mathcal B)), \end{equation} where $\mathcal{Hom}(-,-)$ here denotes the category of ordinary ($k$-linear) functors; $H^0(\mathcal{RHom}(\mathcal A,\mathcal B))$ can be identified to the category of quasi-functors $\mathrm{rep}(\mathcal A,\mathcal B)$.

I've been studying the properties of this functor for a while. For example, I know that for $\mathcal A = \mathbf{1}$, the category with one object and endomorphism ring $k$, $H^0(-)$ is an equivalence; for $\mathcal A = \Delta^1$, the category freely generated over $k$ by the diagram $0 \to 1$, this functor is full and reflects isomorphisms. Assuming $\mathcal B$ is pretriangulated and taking $\mathcal A = (\Delta^1)^{\mathrm{pretr}}$ (the pretriangulated hull of $\Delta^1$) the above claim implies that $H^0(-)$ induces, in this particular case, a bijection between the isomorphism classes of objects.

Do you know other situations where the $H^0$ functor exhibits some good properties as above? For example, are there other cases in which it is full and reflects isomorphisms? Or cases in which it induces a bijection between isomorphism classes of objects?

(Also, I was wondering if a similar problem is studied in the framework of $(\infty,1)$-categories.)

share|cite|improve this question

In the framework of $\infty$-categories, I think it is not difficult to see that your claim for $\mathcal{A} = \Delta^1$ holds for all $\infty$-groupoids.

Let $K$ be an $\infty$-groupoid (Kan complex) and $C$ an $\infty$-category (weak Kan complex). The canonical map $C \to N h C$ is an isofibration (or fibration in the Joyal model structure), where $h$ is the homotopy category functor, left adjoint to the nerve functor $N$. This implies that 1) the induced map $$ \underline{\mathrm{Hom}}(K, C) \longrightarrow \underline{\mathrm{Hom}}(K, NhC) \times_{NhC} C $$ is a trivial Kan fibration, and 2) the canonical map $$ \underline{\mathrm{Hom}}(K, NhC) \times_{NhC} C \longrightarrow \underline{\mathrm{Hom}}(K, NhC) $$ induces a full, conservative, bijective-on-objects functor on the homotopy categories. Since $K$ is a Kan complex, one can prove that $$ N \underline{\mathrm{Hom}}_{\mathrm{Cat}}(hK, hC) = \underline{\mathrm{Hom}}(NhK, NhC) \longrightarrow \underline{\mathrm{Hom}}(K, NhC) $$ is a trivial Kan fibration. Now after passing to homotopy categories one gets a full and conservative functor $$ h\underline{\mathrm{Hom}}(K, C) \longrightarrow \underline{\mathrm{Hom}}_{\mathrm{Cat}}(hK, hC). $$

I am not sure whether the assumption on $K$ Is really necessary.

share|cite|improve this answer

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.